A method, system, device and medium for analyzing performance of network element boards in the entire network

Abstract

A method, system, device and medium for performance analysis of network element boards in the entire network, relating to the field of network communications. In this method, entities in the network structure are divided according to preset classification standards, and a data model of a tree structure is constructed according to the division results; performance indicator data of all network element boards are collected in real time, and the performance indicator data are classified into corresponding tree structure nodes; a preset prediction model is used to dynamically generate a current threshold value according to the current data in the network structure, and whether the performance indicator data of the network element board is normal is determined according to the current threshold value; when there is an abnormal performance indicator data of a network element board, the target network element board is determined, and the abnormality is located according to the path of the target network element board in the tree structure. Implementation of the technical solution provided by the present application can quickly and accurately identify and locate performance abnormalities of network element boards in the entire network, improve fault response speed, and reduce network operation and maintenance costs.

Description

Technical Field

[0001] The present application relates to the technical field of network communications, and specifically to a method, system, device and medium for analyzing the performance of network element boards in the entire network. Background Art

[0002] As the scale of the network continues to expand, performance monitoring of network element boards across the entire network has become one of the key links in ensuring the safe and stable operation of the network.

[0003] Traditional monitoring methods usually rely on manual regular data collection and analysis, which is inefficient and prone to missing abnormal situations. With the continuous expansion of communication networks and the evolution of technology, the number of network devices has increased dramatically, and the complexity of interconnection between devices at all levels has also increased. Traditional manual analysis methods are often inefficient when facing massive data in large-scale networks and are difficult to meet real-time requirements.

[0004] Therefore, more efficient technical means are urgently needed to perform performance analysis and fault location on network element boards across the entire network. Summary of the invention

[0005] The present application provides a method, system, device and medium for analyzing the performance of network element boards in the entire network, which can quickly and accurately identify and locate performance anomalies of network element boards in the entire network, improve fault response speed, and reduce network operation and maintenance costs.

[0006] In a first aspect of the present application, a method for analyzing the performance of network element boards in the entire network is provided, which is applied to a performance analysis platform for network element boards in the entire network. The method includes:

[0007] Divide the entities in the network structure according to the preset classification standard, and construct a tree-structured data model according to the division result, wherein the entities include a computer room, a switch and a network element board;

[0008] Collect performance indicator data of all network element boards in real time and classify the performance indicator data into corresponding tree structure nodes;

[0009] Dynamically generate a current threshold value according to current data in the network structure using a preset prediction model, and determine whether the performance indicator data of the network element board is normal according to the current threshold value;

[0010] When the performance indicator data of a network element board is abnormal, a target network element board is determined, and the abnormality is located according to the path of the target network element board in the tree structure.

[0011] By adopting the above technical solution, the entities in the network structure (such as computer rooms, switches and network element boards) are divided according to the preset classification standards and constructed into a tree-structured data model. This method makes the data management of the network structure clearer and more orderly. The tree structure can intuitively display the hierarchical relationship and subordinate relationship between entities, which is easy to understand and operate. The performance indicator data of all network element boards are collected in real time and classified into the corresponding tree structure nodes, so that the system can grasp the performance status of network equipment in real time. This real-time monitoring capability is crucial for timely discovery and resolution of network problems, and helps to improve the stability and reliability of the network. The current threshold is dynamically generated according to the current data in the network structure using a preset prediction model. This approach is more flexible and accurate than the traditional fixed threshold setting. The dynamic threshold can better adapt to the changes in the network status, so as to more accurately judge whether the performance indicator data of the network element board is normal, and reduce false positives and false negatives. When the performance indicator data of the network element board is abnormal, the target network element board can be determined and located according to its path in the tree structure. This positioning method can quickly locate the problem and improve the efficiency of troubleshooting. At the same time, because the tree structure clearly displays the hierarchical relationship between entities, it can also help operation and maintenance personnel better understand the context in which the exception occurs, so as to more accurately determine the cause and solution of the problem.

[0012] Optionally, dividing the entities in the network structure according to a preset classification standard includes:

[0013] The top layer of the network structure is used as a root node, and the network under the root node is divided according to the geographical location, so that each computer room or data center serves as an independent first-level node;

[0014] Divide the devices under the first-level node according to function or device type, so that each device serves as an independent second-level node;

[0015] The network element boards under the secondary node are divided according to specific components of the device, so that each network element board serves as an independent third-level node.

[0016] By adopting the above technical solution, the top layer of the network structure is taken as the root node, and the network under the root node is divided according to the geographical location, so that each computer room or data center becomes an independent first-level node, which helps to grasp the layout and distribution of the entire network from a macro perspective. The devices under the first-level node are divided according to the function or device type, so that each device becomes an independent second-level node, which helps to understand and analyze the roles and responsibilities of different devices in the network. The network element boards under the second-level node are divided according to the specific components of the device, so that each network element board becomes an independent third-level node, which helps to go deep into the inside of the device and understand the specific composition and operation status of the device. The hierarchical network structure makes data management more orderly and efficient. By building a tree-structured data model, nodes at all levels can be easily managed and operated, such as adding, deleting, and modifying node information. Since each node has a unique identifier and attribute information (such as name, location, function and performance parameters, etc.), this makes data retrieval and query faster and more accurate. When a network failure occurs, the specific location of the failure can be quickly located through the hierarchical network structure. This fault location method not only improves the efficiency of troubleshooting, but also reduces the possibility of misjudgment and missed judgment. The hierarchical network structure makes it possible to flexibly schedule resources. According to the network demand and load, the configuration and resource allocation of each node can be easily adjusted. For example, in some cases, it may be necessary to allocate the resources of a certain computer room or data center to other computer rooms or data centers. In this case, this can be achieved by modifying the data model in the tree structure.

[0017] Optionally, the data model of the tree structure is constructed according to the division result, including:

[0018] Setting a unique identifier for each node and setting corresponding attribute information according to the type and function of the node, wherein the attribute information includes name, location, function and performance parameters;

[0019] According to the hierarchical relationship of the network structure, a parent-child relationship is established for each node, and a data model of a tree structure is constructed according to the parent-child relationship.

[0020] By adopting the above technical solution, a unique identifier is set for each node, ensuring that each entity in the network (such as the computer room, equipment, and network element board) can be accurately identified and located. This is particularly important when querying, updating, or deleting data, and can greatly improve the accuracy of data management. According to the type and function of the node, the corresponding attribute information, such as name, location, function, and performance parameters, is set. These attribute information not only provides a detailed description of the node, but also provides rich data support for subsequent performance analysis, troubleshooting, and resource scheduling. At the same time, the structured storage of attribute information makes data management more efficient and can quickly respond to various network operation requirements. By establishing a parent-child relationship for each node according to the hierarchical relationship of the network structure, and building a tree-structured data model based on this, the network structure becomes intuitive and easy to understand. This tree structure clearly shows the association between each level and entity in the network, which helps operation and maintenance personnel quickly grasp the overall layout and operation status of the network. The intuitive tree structure also simplifies the process of network troubleshooting. When a network failure occurs, the operation and maintenance personnel can troubleshoot step by step along the tree structure and quickly locate the problem, thereby improving the efficiency and accuracy of fault handling. The performance parameter attributes in the tree-structured data model provide a rich data source for performance analysis. By real-time monitoring and analysis of these performance parameters, you can fully understand the network's operating status and performance bottlenecks. The tree structure also makes performance analysis more accurate. Since each node has a clear parent-child relationship and attribute information, the root cause of the performance problem can be accurately located and corresponding optimization measures can be taken.

[0021] Optionally, the classifying the performance indicator data into corresponding tree structure nodes includes:

[0022] Establishing a first mapping rule between performance indicator data and third-level nodes, establishing a second mapping rule between performance indicator data and second-level nodes, and establishing a third mapping rule between performance indicator data and first-level nodes;

[0023] Performing a first classification on the performance indicator data according to the first mapping rule;

[0024] When there is first remaining data in the performance indicator data after the first classification is completed, performing a second classification on the first remaining data according to the second mapping rule, and performing a third classification on the first target data after the second classification is performed according to the tree structure;

[0025] When the first remaining data has second remaining data after the second classification is completed, the second remaining data is fourthly classified according to the third mapping rule, and the second target data that has been subjected to the fourth classification is fifthly classified according to the tree structure.

[0026] By adopting the above technical solution, mapping rules between performance indicator data and third-level nodes, second-level nodes and first-level nodes are established, ensuring that each performance indicator data can be accurately classified to its corresponding node. This mapping rule not only considers the direct attribution of performance indicator data (i.e., the first mapping rule), but also considers the indirect attribution of data (i.e., the second and third mapping rules), thereby ensuring the accuracy of data classification. Through the multi-level classification process, that is, first the first classification, then the second classification of the remaining data, and finally the fourth classification of the remaining data, and subsequent classification according to the tree structure, the integrity of the performance indicator data is ensured. Even if some data cannot find a directly corresponding node in the first classification, it can find a suitable attribution in the subsequent classification, avoiding data loss. The multi-level classification process is actually a divide-and-conquer strategy, which improves the efficiency of data processing by decomposing complex classification problems into multiple simple sub-problems. Each level of classification is based on specific mapping rules, which can be flexibly adjusted and optimized according to actual needs, thereby improving the flexibility of data processing. In addition, the multi-level classification process can also be carried out in parallel, that is, the classification tasks of multiple performance indicator data are processed at the same time, further improving the efficiency of data processing. By classifying the performance indicator data into the corresponding tree structure nodes, we can have a deeper understanding of the performance status of each level and entity of the network. This classification method not only provides an overview of the overall performance of the network, but also reveals the associations and mutual influences between the various components within the network. Since each node has detailed attribute information (such as name, location, function and performance parameters, etc.), a more comprehensive analysis of the performance indicator data can be performed. This comprehensiveness not only helps to discover potential performance bottlenecks and root causes of problems, but also provides strong data support for subsequent optimization measures. When a network failure occurs, by viewing the performance indicator data classified to the corresponding node, the specific location of the failure can be quickly located. Since the data has been classified and organized according to the tree structure, it can be checked step by step along the tree structure until the node or component with the failure is found. This troubleshooting method not only improves accuracy, but also greatly shortens the troubleshooting time. Operation and maintenance personnel can respond and handle failures quickly to ensure the stable operation of the network and a good user experience.

[0027] Optionally, the dynamically generating a current threshold value according to current data in the network structure using a preset prediction model includes:

[0028] Collecting first historical performance data from the network structure, and marking the first historical performance data as normal or abnormal according to the state of the network structure to form a training sample library;

[0029] Using a support vector machine algorithm to train the training sample library to obtain a preliminary prediction model;

[0030] Continuously collecting second historical performance data and evaluating the preliminary prediction model according to the second historical performance data, adjusting the parameters of the support vector machine algorithm according to the evaluation result to train the preset prediction model, wherein the generation time period of the second historical performance data is later than the generation time period of the first historical performance data;

[0031] Current performance data is collected, and current thresholds of different types of performance data are generated based on the current performance data using the preset prediction model.

[0032] By adopting the above technical solution, features are extracted from historical performance data and marked as normal or abnormal, forming a rich training sample library. These sample libraries contain performance data of the network under different states, providing a reliable data basis for training prediction models. By training the training sample library using the support vector machine algorithm, a preliminary prediction model that can accurately distinguish normal and abnormal performance can be obtained. By continuously collecting new historical performance data and evaluating and adjusting the model, the performance of the model can be further optimized to make it more accurately reflect the actual state of the network. Based on the current performance data, the preset prediction model can dynamically generate current thresholds for different types of performance data. These thresholds are calculated based on the current state and trend of the network, so they are more in line with the actual situation of the network and improve the accuracy of threshold setting. The process of dynamically generating current thresholds is real-time, and the thresholds can be adjusted in time according to changes in network status. This means that when network performance is abnormal, the system can respond quickly and issue an alarm, thereby improving the real-time nature of network monitoring. By dynamically generating current thresholds, the system can more accurately monitor changes in network performance. When performance data exceeds the threshold, the system can trigger the corresponding alarm and response mechanism, thereby optimizing network performance in a timely manner. In addition, dynamic thresholds can also help operation and maintenance personnel better understand the resource utilization of the network. By monitoring the threshold changes of different performance indicators, resource bottlenecks and waste can be discovered in a timely manner, and corresponding measures can be taken to optimize and adjust.

[0033] Optionally, locating the abnormality according to the path of the target network element board in the tree structure includes:

[0034] Starting from the target network element board, searching for each target node level by level upward along the parent-child relationship of the tree structure;

[0035] Obtaining a change trend of performance indicators between the target node and the child node of the target node, wherein the performance indicators include CPU usage, memory occupancy, and network throughput;

[0036] Determine whether the target node is the root cause of the anomaly based on the change trend, the associated anomalies with other nodes, and the historical performance of the target node.

[0037] By adopting the above technical solution, starting from the target network element board, each target node is searched up step by step along the parent-child relationship of the tree structure, and the source of the abnormality can be systematically traced. This method not only considers the directly related nodes, but also considers the upper-level nodes that may affect the target network element board, thereby improving the accuracy of abnormal location. In the positioning process, the performance indicator change trend between the child node of the target node and the target node, such as CPU usage, memory occupancy, and network throughput, is obtained, providing rich data support for abnormal analysis. These performance indicators can intuitively reflect the operating status of the node and help discover potential performance bottlenecks or abnormal modes. According to the change trend of the performance indicators, the associated abnormalities with other nodes, and the historical performance of the target node, it is comprehensively judged whether the target node is the root cause of the abnormality. This comprehensive judgment method not only considers the current performance indicator data, but also combines historical data and the association relationship between nodes, improving the accuracy and reliability of abnormal location.

[0038] Optionally, the method further includes:

[0039] According to the flow distribution of the network structure, nodes with flow exceeding a preset threshold are taken as target nodes;

[0040] On the target node, deploying one or more communication interfaces as redundant interfaces;

[0041] The state of the primary interface of the target node is detected by a heartbeat packet, and when a fault signal is detected, the data channel is switched from the primary interface to a redundant interface for data transmission.

[0042] By adopting the above technical solution, according to the traffic distribution of the network structure, nodes with more traffic than the preset threshold are taken as target nodes, and the key nodes in the network can be accurately identified. Since these nodes carry a large number of data transmission tasks, they are more likely to become potential risk points of network failure. Deploying one or more communication interfaces as redundant interfaces on the target node provides additional channels for data transmission. When the main interface fails, the system can quickly switch the data channel to the redundant interface to ensure the continuity and stability of data transmission. The status of the main interface of the target node can be detected by the heartbeat packet, and the working condition of the main interface can be monitored in real time. Once a fault signal is detected, the system can immediately start the switching mechanism to switch the data channel from the main interface to the redundant interface. This fault switching mechanism is automatic and fast, and can be completed without interrupting data transmission. Therefore, it significantly enhances the fault recovery capability of the network and reduces the risk of business interruption caused by network failure.

[0043] In a second aspect of the present application, a system for analyzing the performance of network element boards in the entire network is provided, including a structure module, a classification module, a judgment module and a positioning module, wherein:

[0044] A structure module is configured to divide entities in the network structure according to a preset classification standard and construct a tree-structured data model according to the division result, wherein the entities include a computer room, a switch and a network element board;

[0045] A classification module is configured to collect performance indicator data of all network element boards in real time and classify the performance indicator data into corresponding tree structure nodes;

[0046] A judgment module configured to dynamically generate a current threshold value according to current data in the network structure using a preset prediction model, and determine whether the performance indicator data of the network element board is normal according to the current threshold value;

[0047] The positioning module is configured to determine the target network element board when the performance indicator data of the network element board is abnormal, and locate the abnormality according to the path of the target network element board in the tree structure.

[0048] In the third aspect of the present application, an electronic device is provided, including a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes any one of the methods described above.

[0049] In a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores instructions, and when the instructions are executed, any of the methods described above is executed.

[0050] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0051] 1. By dividing the entities in the network structure (such as computer rooms, switches, and network element boards) according to preset classification standards and building a tree-structured data model, hierarchical and structured management of the network structure is achieved. This method helps to clearly display the network structure and facilitates operation and maintenance personnel to understand and monitor;

[0052] 2. Collect the performance indicator data of all network element boards in real time and classify them into corresponding tree structure nodes to ensure the real-time and accuracy of the data. This helps operation and maintenance personnel to understand the network performance status in a timely manner and discover potential problems;

[0053] 3. Use the preset prediction model to dynamically generate the current threshold value based on the current data in the network structure. This method is more flexible and accurate than the traditional static threshold setting. Through the dynamic threshold value, it is possible to more accurately determine whether the performance indicator data of the network element board is normal, so as to detect abnormalities in time;

[0054] 4. When the performance indicator data of a network element board is abnormal, the specific location of the abnormality can be quickly found by determining the target network element board and locating the abnormality according to its path in the tree structure. This method helps operation and maintenance personnel quickly locate the root cause of the problem and reduce troubleshooting time;

[0055] 5. It combines multiple links such as real-time data collection, dynamic threshold setting and accurate anomaly positioning to form a complete performance analysis process. This helps improve operation and maintenance efficiency, reduce human errors, and improve the intelligence level of network operation and maintenance;

[0056] 6. By timely discovering and handling the performance anomalies of network element boards, the problem can be prevented from further deteriorating, thereby ensuring the stability and performance of the network. In addition, it can also provide data support for network optimization and help operation and maintenance personnel make more scientific decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a flow chart of the method for analyzing the performance of network element boards in the entire network disclosed in the embodiment of the present application;

[0058] Figure 2 It is a module schematic diagram of a system for analyzing performance of network element boards in the entire network disclosed in an embodiment of the present application;

[0059] Figure 3 It is a structural schematic diagram of an electronic device disclosed in an embodiment of the present application.

[0060] Explanation of the reference numerals: 201, structural module; 202, classification module; 203, judgment module; 204, positioning module; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. DETAILED DESCRIPTION

[0061] In order to enable technicians in this field to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments.

[0062] In the description of the embodiments of the present application, words such as "for example" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "for example" 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 "for example" or "for example" is intended to present related concepts in a specific way.

[0063] In the description of the embodiments of the present application, the meaning of the term "multiple" refers to two or more. For example, multiple systems refer to two or more systems, and multiple screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized.

[0064] This embodiment discloses a method for analyzing the performance of network element boards in the entire network, which is applied to a performance analysis platform for network element boards in the entire network. Figure 1 is a flow chart of the method for analyzing the performance of network element boards in the entire network disclosed in the embodiment of the present application, such as Figure 1 As shown, the method comprises the following steps:

[0065] S110, dividing entities in the network structure according to a preset classification standard, and constructing a tree-structured data model according to the division result, wherein the entities include a computer room, a switch, and a network element board;

[0066] A network element (NE) is the smallest unit that can be monitored and managed in network management. It consists of one or more disks or frames and can independently complete certain transmission functions. In the network management system, a network element can be understood as an element or device in the network, such as PDH equipment, and more commonly, base stations and wireless controllers. These network elements together constitute a communication network and provide infrastructure for communication services. In a communication network, a single board usually refers to a component in a network element. It is a circuit board with specific functions that is used to implement certain transmission or processing functions of the network element. A single board usually contains a processor, memory, interface circuit, etc. to support the implementation of its functions. Different network elements may contain different types and quantities of single boards to meet different communication needs. The network element single board of the entire network refers to the general term for the single boards contained in all network elements in the entire communication network. These single boards jointly realize the transmission, processing and control functions of the communication network through connection and collaboration. In a communication network, each network element is an independent entity, and the single board is an important part of these entities. Therefore, the status and performance of the network element single board of the entire network directly affects the stability and reliability of the entire communication network.

[0067] The computer room is the core part of the network infrastructure. It provides the physical environment for network devices, servers, etc. to operate. In the network structure, the computer room can be regarded as an independent layer or node because it contains multiple network devices and services. A switch is a type of network device that is used to connect multiple devices such as computers, servers, etc. in a local area network. It forwards data to the correct port based on the destination address of the data packet. In the network structure, the switch is usually located inside the computer room as a key node for data communication. The network element board is one of the basic units that constitute network devices such as switches and routers. It contains hardware and software components for processing data packets and is responsible for implementing specific functions of network devices. In the network structure, the network element board is the core component that constitutes network devices such as switches.

[0068] Based on the above entity division, a tree-structured data model can be constructed to organize these entities. The tree structure is a hierarchical data structure that is very suitable for representing entities with hierarchical relationships. In this model, the top layer of the network structure can be used as the root node because it contains the physical foundation of the entire network structure. Based on the root node, the computer room, switch, and network element board can be added to the tree structure in order according to the hierarchy. If the tree structure needs to be represented in more detail, more child nodes can be added under the network element board node, such as different hardware components or software modules. In the tree structure, each node is connected to its child nodes through connections (or edges). These connections represent the hierarchical relationship between entities.

[0069] Optionally, dividing the entities in the network structure according to a preset classification standard includes:

[0070] The top layer of the network structure is used as a root node, and the network under the root node is divided according to the geographical location, so that each computer room or data center serves as an independent first-level node;

[0071] Divide the devices under the first-level node according to function or device type, so that each device serves as an independent second-level node;

[0072] The network element boards under the secondary node are divided according to specific components of the device, so that each network element board serves as an independent third-level node.

[0073] The top layer of the network structure is considered the root node of the entire model. This root node represents the starting point and the highest level of the entire network structure. It contains all entities in the network and the connections between them. Under the root node, the network is divided according to geographical location. This usually means dividing the network into different geographical areas or locations, such as different cities, regions or countries. Each computer room or data center is an independent first-level node, representing the network infrastructure of that geographical location. This division method helps to understand the distribution of the network in different geographical locations, as well as the network connection and communication between different geographical locations. At the same time, it also facilitates subsequent troubleshooting, performance monitoring and resource allocation. Under each first-level node (computer room or data center), the devices are further divided according to function or device type. This usually means dividing the devices into different categories, such as switches, routers, servers, etc. Each device is an independent second-level node, representing a specific device instance under this category. This division method helps to understand the functions and roles of different devices in the network, as well as the connection and communication relationships between them. At the same time, it also facilitates subsequent device management, configuration and upgrade. Under each second-level node (device), the network element boards are further divided according to the specific components of the device. The NE board is one of the core components of network equipment. It contains hardware and software components for processing data packets. Each NE board is an independent third-level node, representing a specific component instance under the device. This division method helps to gain a deeper understanding of the internal structure and components of the device, as well as their functions and roles in the network. At the same time, it also provides more detailed information and basis for subsequent troubleshooting, performance monitoring and component upgrades.

[0074] By taking the top layer of the network structure as the root node and dividing it level by level according to geographical location, function or device type, and specific components of the device, the entire network structure presents a clear hierarchical relationship. This hierarchical representation method helps operation and maintenance personnel better understand and grasp the overall architecture of the network. When a network failure occurs, the specific machine room, device, or network element board can be quickly located based on the fault phenomenon. This accurate fault location method helps to reduce the time of troubleshooting and improve the efficiency of fault handling. At the same time, since each node contains detailed device information, operation and maintenance personnel can also conduct more in-depth analysis and diagnosis of the fault based on this information.

[0075] Optionally, the data model of the tree structure is constructed according to the division result, including:

[0076] Setting a unique identifier for each node and setting corresponding attribute information according to the type and function of the node, wherein the attribute information includes name, location, function and performance parameters;

[0077] According to the hierarchical relationship of the network structure, a parent-child relationship is established for each node, and a data model of a tree structure is constructed according to the parent-child relationship.

[0078] When building a tree-structured data model, in order to distinguish different nodes, a unique identifier needs to be set for each node. This identifier should be unique in the entire network structure so that a specific node can be accurately located in subsequent operations and queries. In addition to the unique identifier, corresponding attribute information needs to be set for each node to describe the node's type, function, location, and performance parameters. The name of the node can be a concise and clear identifier for display in the user interface or log. For computer room or data center nodes, location information may include geographic location (such as city, region) and physical location (such as floor, room number). For equipment and network element single board nodes, location information may include their specific location in the computer room or equipment (such as rack number, slot number). The function of the node describes its role in the network structure, such as router, switch, server, etc. For network element single board nodes, the function may be more specific, such as transmission board, processing board, etc. The performance parameters of the node may include processing power, storage capacity, transmission rate, etc., which help evaluate the performance and capacity of the node.

[0079] In the network structure, there is a certain hierarchical relationship between nodes. For example, the computer room or data center is the highest-level node, the equipment is the second-highest-level node, and the network element board is the lower-level node. This hierarchical relationship reflects the organization and functional division of the network structure. According to the hierarchical relationship of the network structure, a parent-child relationship can be established for each node. For example, a computer room node can have multiple device nodes as its child nodes, and a device node can have multiple network element board nodes as its child nodes. This parent-child relationship forms the basis of the tree structure. The data model of the tree structure is a hierarchical data structure in which each node has one or more child nodes, but only one parent node (except the root node). This data structure can clearly represent the hierarchical relationship and parent-child relationship between nodes. When constructing a tree structure data model, first, all nodes need to be organized according to the hierarchical relationship. Then, a unique identifier and attribute information are added to each node. Finally, the nodes are connected according to the parent-child relationship to form a complete tree structure.

[0080] By setting a unique identifier for each node and setting corresponding attribute information according to the type and function of the node, the orderly organization and structured storage of data are achieved. This structured data model makes network data easier to manage and query, and improves the efficiency of data processing. The setting of node attribute information includes multiple aspects such as name, location, function and performance parameters, which together constitute a complete description of the node. This comprehensive information record helps operation and maintenance personnel to better understand the status of network nodes and provides rich data support for network management and maintenance. According to the hierarchical relationship of the network structure, a parent-child relationship is established for each node, and a tree-structured data model is constructed accordingly. This hierarchical data representation method makes the network structure more intuitive and easy to understand, which helps operation and maintenance personnel to better grasp the overall architecture of the network and the relationship between each node. The structured data model and clear hierarchical relationship facilitate complex query and analysis. Operation and maintenance personnel can traverse and calculate the data model through query statements or analysis algorithms according to actual needs to obtain the required information and results. This capability helps to achieve deep mining and intelligent analysis of network data, providing strong support for network optimization and decision-making.

[0081] S120, collecting performance indicator data of all network element boards in real time, and classifying the performance indicator data into corresponding tree structure nodes;

[0082] Performance indicator data usually comes from the network element board itself or the management system connected to it. This data may include key performance indicators such as CPU usage, memory usage, disk space, network throughput, packet loss rate, and latency. Collection methods may include polling, event-driven, or agent-based collection. The polling method refers to sending requests to the network element board regularly to obtain its performance indicator data. The event-driven method refers to actively sending notifications to the management system when a specific event (such as performance abnormality) occurs on the network element board. The agent-based method refers to installing agent software on the network device, which is responsible for collecting and sending performance indicator data. The collected performance indicator data is usually transmitted and stored in a specific format (such as XML, JSON, SNMP, etc.). These formats ensure the readability and parsability of the data. The classification method is usually based on the identifier of the network element board (such as IP address, MAC address or device ID) and the hierarchical relationship of the tree structure nodes. First, the system needs to determine the network element board to which the collected performance indicator data belongs. Then, according to the position of the network element board in the tree structure, its performance indicator data is classified into the corresponding node. The classified performance indicator data can be stored in a database, memory cache or distributed storage system. These storage systems ensure the persistence and accessibility of the data. At the same time, in order to support real-time analysis and monitoring, the data usually needs to be indexed and queried in an efficient way.

[0083] Optionally, the classifying the performance indicator data into corresponding tree structure nodes includes:

[0084] Establishing a first mapping rule between performance indicator data and third-level nodes, establishing a second mapping rule between performance indicator data and second-level nodes, and establishing a third mapping rule between performance indicator data and first-level nodes;

[0085] Performing a first classification on the performance indicator data according to the first mapping rule;

[0086] When there is first remaining data in the performance indicator data after the first classification is completed, performing a second classification on the first remaining data according to the second mapping rule, and performing a third classification on the first target data after the second classification is performed according to the tree structure;

[0087] When the first remaining data has second remaining data after the second classification is completed, the second remaining data is fourthly classified according to the third mapping rule, and the second target data that has been subjected to the fourth classification is fifthly classified according to the tree structure.

[0088] The first mapping rule defines the correspondence between performance indicator data and third-level nodes (i.e., network element board nodes). Each performance indicator data item should be uniquely mapped to a third-level node, indicating that the data item is generated by the network element board. The second mapping rule defines the correspondence between the first remaining data and the second-level node (i.e., device node) when the first mapping rule cannot fully classify the performance indicator data (i.e., when the first remaining data exists). This means that if a certain performance indicator data cannot be directly associated with a specific network element board, then it should be classified into the device to which the network element board belongs. The third mapping rule defines the correspondence between the second remaining data and the first-level node (i.e., the computer room or data center node) when the second mapping rule cannot fully classify the first remaining data (i.e., the second remaining data still exists). This means that if a certain performance indicator data cannot be associated with a specific network element board or a specific device, then it should be classified into the computer room or data center where the device is located.

[0089] According to the first mapping rule, the performance indicator data that can be directly associated with a specific network element board are classified. These data will be classified into the corresponding third-level nodes. After the first classification is completed, if there is residual data that cannot be directly associated with a specific network element board (i.e., the first residual data), these data are classified into the corresponding second-level nodes according to the second mapping rule. This step is to ensure that the data that cannot be directly classified into the network element board can be classified into a higher-level node. It should be noted that when performing the second classification, the data that has been classified into the second-level node (i.e., the first target data) may be further classified (i.e., the fourth classification) or verified to ensure the integrity and accuracy of the data. This classification or verification may be based on the hierarchical relationship of the tree structure to ensure that the transmission and inheritance of data between nodes are correct. After the second classification is completed, if there is still residual data that cannot be associated with a specific device (these data may actually be unable to determine the ownership from the device level), these data are classified into the corresponding first-level nodes according to the tree structure and the third mapping rule. This step is to ensure that the data that cannot be determined at the lower level can be classified into the first-level node. Likewise, when executing the third classification, the data that has been classified into the first-level node may be further classified (ie, the fifth classification) or verified.

[0090] By establishing multi-level mapping rules (first, second, and third mapping rules), refined management and monitoring of performance indicator data is achieved. This refinement is not only reflected in the accuracy of data classification, but also in the ability to flexibly adjust and manage performance indicator data according to the needs of nodes at different levels. A hierarchical classification method (first classification, second classification, and fourth classification) and further classification based on a tree structure (third classification and fifth classification) are adopted. This hierarchical processing method helps to improve the efficiency of data classification. By first processing the most directly related data (i.e., third-level node data) and then gradually expanding to a wider range of data (secondary and primary node data), the accuracy and timeliness of data classification can be ensured. Considering the complexity and diversity of performance indicator data, this process allows for residual data after the first or second classification. These residual data can be further classified by a higher-level mapping rule to ensure that all data can be correctly classified into the corresponding tree structure node. This flexibility helps to deal with performance indicator data from different sources and in different formats. By accurately classifying performance indicator data into corresponding tree structure nodes, the performance status and resource usage of network devices can be more clearly understood. This helps optimize resource utilization and avoid resource waste. At the same time, accurate data classification also provides more reliable decision support for operation and maintenance personnel, helping to formulate targeted maintenance and management strategies.

[0091] S130, dynamically generating a current threshold value according to current data in the network structure using a preset prediction model, and determining whether the performance indicator data of the network element board is normal according to the current threshold value;

[0092] The selection of the preset prediction model should be based on the characteristics of the network performance data and business requirements. Commonly used models include time series prediction models (such as ARIMA, SARIMA), machine learning models (such as decision trees, random forests, support vector machines, neural networks, etc.), and deep learning models (such as recurrent neural networks, long short-term memory networks, etc.). Before building the model, the historical performance data in the network structure needs to be preprocessed. This includes steps such as data cleaning (removing outliers, filling missing values, etc.), data transformation (such as normalization, standardization, etc.), and feature selection (extracting features that have a significant impact on network performance). The preset prediction model is trained using the preprocessed historical performance data. During the training process, the parameters of the model need to be adjusted to optimize its prediction performance. This usually involves strategies such as cross-validation and grid search to find the best combination of model parameters. After training, the model needs to be validated to evaluate its prediction performance. This can be achieved by comparing the difference between the model prediction value and the actual value. Common evaluation indicators include mean square error (MSE), root mean square error (RMSE), mean absolute error (MAE), etc. Collect current data from the network structure in real time, including performance indicator data of network element boards (such as CPU usage, memory usage, network throughput, etc.). Input the real-time collected data into the preset prediction model to generate predicted values ​​for performance indicator data in the future. Calculate the current threshold based on the predicted value. This can be achieved by setting a certain confidence interval or determining a reasonable range based on business needs. For example, the 95% confidence interval of the predicted value can be used as the threshold for anomaly detection. Compare the real-time collected network element board performance indicator data with the current threshold. If the performance indicator data exceeds the threshold range, the data is considered abnormal; otherwise, the data is considered normal. When abnormal data is detected, the alarm mechanism can be triggered to notify the operation and maintenance personnel to handle it. At the same time, the abnormal data can also be recorded for subsequent analysis and diagnosis.

[0093] Optionally, the dynamically generating a current threshold value according to current data in the network structure using a preset prediction model includes:

[0094] Collecting first historical performance data from the network structure, and marking the first historical performance data as normal or abnormal according to the state of the network structure to form a training sample library;

[0095] Using a support vector machine algorithm to train the training sample library to obtain a preliminary prediction model;

[0096] Continuously collecting second historical performance data and evaluating the preliminary prediction model according to the second historical performance data, adjusting the parameters of the support vector machine algorithm according to the evaluation result to train the preset prediction model, wherein the generation time period of the second historical performance data is later than the generation time period of the first historical performance data;

[0097] Current performance data is collected, and current thresholds of different types of performance data are generated based on the current performance data using the preset prediction model.

[0098] The performance data of network element boards over a period of time (such as the past few months or years) are collected from the network structure. These data constitute the first historical performance data set. These data may include various types of performance indicators such as CPU usage, memory usage, disk I / O, network throughput, etc. The first historical performance data is labeled as normal or abnormal according to the status of the network structure (such as whether a fault has occurred, whether maintenance is being performed, etc.) and business needs. This usually involves in-depth analysis and understanding of the data to ensure the accuracy of the labeling. The labeled data forms a training sample library. The support vector machine (SVM) algorithm is used as the basis of the prediction model. SVM is a commonly used machine learning algorithm that is particularly suitable for handling classification problems. The labeled training sample library is input into the SVM algorithm for model training. During the training process, the SVM algorithm learns how to distinguish between normal and abnormal performance data and finds an optimal decision boundary (i.e., a hyperplane) to maximize the interval between the two types of data. After the training of the preliminary prediction model is completed, new historical performance data (i.e., second historical performance data) continues to be collected. The generation time period of the second historical performance data should be later than the generation time period of the first historical performance data to ensure the timeliness and accuracy of the model. The second historical performance data is used to evaluate the preliminary prediction model. This usually involves applying the model to new data and comparing the difference between the model's predicted results and the actual results. Evaluation indicators may include accuracy, recall, F1 score, etc. Based on the evaluation results, adjust the parameters of the SVM algorithm (such as C parameters, kernel function parameters, etc.) to optimize the performance of the model. This may require multiple iterations and adjustments until the best parameter combination is found. After multiple iterations and adjustments, the preset prediction model is finally obtained. This model has high accuracy and generalization ability and can handle new performance data well. Collect current performance data from the network structure in real time. This data may include multiple types of performance indicators, such as CPU usage, memory usage, etc. Use the preset prediction model to generate current thresholds for different types of performance data based on the current performance data. This usually involves inputting the current performance data into the model and determining a reasonable threshold range based on the output of the model. The selection of thresholds should be based on business needs and the actual situation of the network structure.

[0099] By collecting the first historical performance data and marking it as normal or abnormal, a training sample library is constructed. This step ensures that the data basis for model training is accurate and representative. The training sample library is trained using the support vector machine algorithm to obtain a preliminary prediction model. Further, the second historical performance data is continuously collected to evaluate the preliminary prediction model, and the model parameters are adjusted according to the evaluation results, and finally a preset prediction model is obtained. This continuous learning and adjustment process enables the model to more accurately reflect the dynamic changes of network performance data, thereby generating more accurate current thresholds. The preset prediction model can generate current thresholds for different types of performance data based on current performance data. These thresholds are dynamically generated according to the current state of the network structure and the real-time changes of performance data, so they can more accurately reflect the normal range of network performance. When the actual performance data exceeds these thresholds, the anomaly can be discovered and reported in time, which improves the accuracy and timeliness of anomaly detection. By dynamically generating current thresholds and performing anomaly detection based on these thresholds, network operation and maintenance personnel can more accurately locate and solve network performance problems. This avoids unnecessary waste of resources and increased operation and maintenance costs. At the same time, since the model can automatically learn and adjust, the need for manual intervention is reduced, and the operation and maintenance efficiency is improved.

[0100] S140: When performance indicator data of a network element board is abnormal, determine a target network element board, and locate the abnormality according to a path of the target network element board in the tree structure.

[0101] When an anomaly is confirmed, the system immediately identifies the network element board where the anomaly occurred, i.e., the target network element board. This is usually achieved by comparing the device ID, IP address, or other unique identifier to which the anomaly data belongs. When the target network element board is determined, the system searches for its path based on its location in the tree structure. This path starts from the root node of the tree and extends to the leaf node where the target network element board is located. By marking the location of the target network element board in the tree structure, the system can clearly display the specific location where the anomaly occurred. This helps operation and maintenance personnel quickly understand the scope of the anomaly and take appropriate measures to troubleshoot and repair it. In order to improve operation and maintenance efficiency, network management systems usually provide visual interfaces to display the tree structure and the location of the anomaly. Through these interfaces, operation and maintenance personnel can intuitively see the network element board where the anomaly occurred and its location in the network, so as to make decisions faster. Based on the location of the anomaly, the operation and maintenance personnel can start troubleshooting. This may include checking the hardware status, software configuration, network connection, etc. of the network element board. Once the cause of the fault is found, the operation and maintenance personnel can take appropriate repair measures to solve the problem. This may include restarting the device, updating software, replacing hardware, etc. After handling the exception, the operation and maintenance personnel need to record the cause of the exception, the handling process and the results. This information can be used for subsequent network optimization and fault prevention work. At the same time, the exception information and handling experience can also be fed back to the developers or suppliers of the network management system so that they can improve the performance and functions of the system.

[0102] Optionally, locating the abnormality according to the path of the target network element board in the tree structure includes:

[0103] Starting from the target network element board, searching for each target node level by level upward along the parent-child relationship of the tree structure;

[0104] Obtaining a change trend of performance indicators between the target node and the child node of the target node, wherein the performance indicators include CPU usage, memory occupancy, and network throughput;

[0105] Determine whether the target node is the root cause of the anomaly based on the change trend, the associated anomalies with other nodes, and the historical performance of the target node.

[0106] When the network management system detects that the performance indicator data of a certain network element board (i.e., the target network element board) is abnormal, the starting point for locating the abnormality is the target network element board. Starting from the target network element board, the system searches upward step by step along the parent-child relationship (i.e., the relationship between the upper network element and the lower network element) of the tree structure. This usually means starting from the leaf node (i.e., the target network element board) and tracing back to the root node (the entire network). In this process, the system records each node passed as a target node. These target nodes are potential sources of problems because their performance problems may affect downstream child nodes. For each target node, the system collects performance indicator data between its child nodes (including the target network element board) and the target node. These performance indicators usually include CPU usage, memory usage, and network throughput, which can reflect the operating status and performance bottlenecks of the device. The system analyzes the changing trends of these performance indicator data. This includes comparing data values ​​at different time points and calculating the growth rate or decline rate of the data. Through these analyses, the system can identify whether the performance indicators have experienced abnormal fluctuations or continued deterioration. Based on the trend of performance indicators, the system will evaluate whether the target node may be the root cause of the anomaly. For example, if the CPU usage of the target node is continuously too high, and this high usage is directly related to the performance problems of the downstream child nodes, then the target node is likely to be the source of the problem. In addition to analyzing the performance indicators of a single target node, the system will also consider the associated anomalies with other nodes. For example, if multiple child nodes report similar performance problems at the same time, and these problems are all related to a common parent node (i.e., the target node), then this parent node is likely to be the root cause of the problem. The system will also compare the historical performance of the target node. If the target node has also experienced similar performance problems in the past, and these problems are ultimately confirmed to be caused by the failure of the node itself or configuration errors, then this anomaly is also likely to be caused by the same reason. After combining all the above information, the system will make a preliminary judgment: whether the target node is indeed the root cause of the anomaly. If the judgment result is positive, the system can generate a detailed anomaly report indicating the specific location of the problem, possible causes, and recommended repair measures. If the judgment result is negative, the system will continue to search upward along the tree structure path until the true root cause of the anomaly is found.

[0107] By starting from the target network element board and searching for each target node step by step along the parent-child relationship of the tree structure, the specific location where the anomaly occurred can be accurately located. This method not only considers the directly related network element boards, but also considers their hierarchical relationship and mutual connection in the network structure, thereby improving the accuracy of positioning. Obtaining the change trend of performance indicators (such as CPU usage, memory occupancy, and network throughput) between the child nodes of the target node and the target node helps to deeply understand the background and cause of the anomaly. The change trend of these performance indicators can provide important information about the network status before and after the anomaly occurs, which helps operation and maintenance personnel to better understand and analyze the anomaly. Based on the change trend of performance indicators, the associated anomalies with other nodes, and the historical performance of the target node, it can be comprehensively judged whether the target node is the root cause of the anomaly. This method considers multiple factors, including the change of performance indicators, the mutual influence between nodes, and the historical behavior of nodes, thereby improving the accuracy of judgment. Through the above process, operation and maintenance personnel can locate the root cause of the anomaly more quickly, thereby reducing the time for troubleshooting. This helps to improve the efficiency of network operation and maintenance and reduce the risk of service interruption caused by network failures. By automatically monitoring and analyzing the changing trends of performance indicators and automatically locating the root causes of anomalies, the network management system can gradually enhance its self-healing capabilities. This means that the network can automatically take appropriate measures to restore normal operation when an anomaly occurs without human intervention.

[0108] Optionally, the method further includes:

[0109] According to the flow distribution of the network structure, nodes with flow exceeding a preset threshold are taken as target nodes;

[0110] On the target node, deploying one or more communication interfaces as redundant interfaces;

[0111] The state of the primary interface of the target node is detected by a heartbeat packet, and when a fault signal is detected, the data channel is switched from the primary interface to a redundant interface for data transmission.

[0112] Perform a detailed analysis of the traffic distribution of the entire network structure. This usually involves collecting traffic data from each node and performing statistics and comparisons. Set a reasonable traffic threshold based on the network scale, business requirements, and traffic characteristics. The traffic threshold is used to determine whether the traffic of a node requires special attention. Nodes whose traffic exceeds the preset threshold are used as target nodes. These nodes are more likely to become network bottlenecks or failure points due to their high traffic, so redundant configuration is required. On the target node, select one or more communication interfaces as redundant interfaces based on actual needs and network environment. These interfaces can be physical interfaces or virtual interfaces. Perform necessary configuration on the redundant interfaces, including network parameters such as IP address, subnet mask, gateway, and related security policies and routing rules. After deployment, the redundant interfaces need to be tested to ensure that they can work properly and communicate normally with other nodes in the network. In the network, the status of the target node's primary interface is detected by sending heartbeat packets. A heartbeat packet is a data packet in a specific format that is sent periodically to confirm the node's online status and connection quality. When the target node's primary interface fails, the sending or receiving of heartbeat packets will be affected. By monitoring the sending and receiving of heartbeat packets, the failure of the primary interface can be discovered in time. Once the failure of the primary interface is detected, the system will immediately trigger the switching mechanism to switch the data channel from the primary interface to the redundant interface. This switching process needs to ensure the continuity and integrity of data transmission to avoid data loss or interruption.

[0113] By deploying redundant interfaces and heartbeat packet detection mechanisms, network failures can be discovered and responded to in a timely manner to ensure the continuity and stability of data transmission. Deploying redundant interfaces on high-traffic nodes can disperse network traffic, alleviate network bottlenecks, and improve overall network performance. The number and configuration of redundant interfaces can be flexibly adjusted according to changes in network scale and business needs. Through automated detection and switching mechanisms, manual intervention and operation and maintenance costs can be reduced, and operation and maintenance efficiency can be improved.

[0114] This embodiment also discloses a system for analyzing the performance of network element boards in the entire network. Figure 2 Schematic diagram of the module of the system for analyzing the performance of network element boards in the whole network disclosed in the embodiment of the present application, such as Figure 2 As shown, the system includes a structure module 201, a classification module 202, a judgment module 203 and a positioning module 204, wherein:

[0115] The structure module 201 is configured to divide entities in the network structure according to a preset classification standard, and construct a tree-structured data model according to the division result, wherein the entities include a computer room, a switch and a network element board;

[0116] The classification module 202 is configured to collect the performance index data of all network element boards in real time and classify the performance index data into corresponding tree structure nodes;

[0117] The judgment module 203 is configured to dynamically generate a current threshold value according to the current data in the network structure by using a preset prediction model, and determine whether the performance indicator data of the network element board is normal according to the current threshold value;

[0118] The positioning module 204 is configured to determine a target network element board when there is an abnormality in the performance indicator data of the network element board, and locate the abnormality according to the path of the target network element board in the tree structure.

[0119] Optionally, the structural module 201 is configured to:

[0120] The top layer of the network structure is used as a root node, and the network under the root node is divided according to the geographical location, so that each computer room or data center serves as an independent first-level node;

[0121] Divide the devices under the first-level node according to function or device type, so that each device serves as an independent second-level node;

[0122] The network element boards under the secondary node are divided according to specific components of the device, so that each network element board serves as an independent third-level node.

[0123] Optionally, the structural module 201 is configured to:

[0124] Setting a unique identifier for each node and setting corresponding attribute information according to the type and function of the node, wherein the attribute information includes name, location, function and performance parameters;

[0125] According to the hierarchical relationship of the network structure, a parent-child relationship is established for each node, and a data model of a tree structure is constructed according to the parent-child relationship.

[0126] Optionally, the classification module 202 is configured to:

[0127] Establishing a first mapping rule between performance indicator data and third-level nodes, establishing a second mapping rule between performance indicator data and second-level nodes, and establishing a third mapping rule between performance indicator data and first-level nodes;

[0128] Performing a first classification on the performance indicator data according to the first mapping rule;

[0129] When there is first remaining data in the performance indicator data after the first classification is completed, performing a second classification on the first remaining data according to the second mapping rule, and performing a third classification on the first target data after the second classification is performed according to the tree structure;

[0130] When the first remaining data has second remaining data after the second classification is completed, the second remaining data is fourthly classified according to the third mapping rule, and the second target data that has been subjected to the fourth classification is fifthly classified according to the tree structure.

[0131] Optionally, the judging module 203 is configured to:

[0132] Collecting first historical performance data from the network structure, and marking the first historical performance data as normal or abnormal according to the state of the network structure to form a training sample library;

[0133] Using a support vector machine algorithm to train the training sample library to obtain a preliminary prediction model;

[0134] Continuously collecting second historical performance data and evaluating the preliminary prediction model according to the second historical performance data, adjusting the parameters of the support vector machine algorithm according to the evaluation result to train the preset prediction model, wherein the generation time period of the second historical performance data is later than the generation time period of the first historical performance data;

[0135] Current performance data is collected, and current thresholds of different types of performance data are generated based on the current performance data using the preset prediction model.

[0136] Optionally, the positioning module 204 is configured to:

[0137] Starting from the target network element board, searching for each target node level by level upward along the parent-child relationship of the tree structure;

[0138] Obtaining a change trend of performance indicators between the target node and the child node of the target node, wherein the performance indicators include CPU usage, memory occupancy, and network throughput;

[0139] Determine whether the target node is the root cause of the anomaly based on the change trend, the associated anomalies with other nodes, and the historical performance of the target node.

[0140] Optionally, the system further comprises a redundancy module, wherein the redundancy module is configured to:

[0141] According to the flow distribution of the network structure, nodes with flow exceeding a preset threshold are taken as target nodes;

[0142] On the target node, deploying one or more communication interfaces as redundant interfaces;

[0143] The state of the primary interface of the target node is detected by a heartbeat packet, and when a fault signal is detected, the data channel is switched from the primary interface to a redundant interface for data transmission.

[0144] It should be noted that: when the device provided in the above embodiment realizes its function, only the division of the above functional modules is used as an example. In actual application, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be repeated here.

[0145] This embodiment also discloses an electronic device, referring to Figure 3 The electronic device may include: at least one processor 301 , at least one communication bus 302 , a user interface 303 , a network interface 304 , and at least one memory 305 .

[0146] The communication bus 302 is used to realize the connection and communication between these components.

[0147] The user interface 303 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0148] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).

[0149] Among them, the processor 301 may include one or more processing cores. The processor 301 uses various interfaces and lines to connect various parts in the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory 305, and calling data stored in the memory 305. Optionally, the processor 301 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 301 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 301, and it can be implemented separately through a chip.

[0150] Among them, the memory 305 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 305 includes a non-transitory computer-readable storage medium. The memory 305 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 305 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 305 may also be optionally at least one storage device located away from the aforementioned processor 301. As Figure 3 As shown, the memory 305 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an application program of a method for analyzing the performance of network element boards in the entire network.

[0151] exist Figure 3In the electronic device shown, the user interface 303 is mainly used to provide an input interface for the user and obtain data input by the user; and the processor 301 can be used to call the application program of the method of performance analysis of network element boards in the entire network stored in the memory 305. When executed by one or more processors 301, the electronic device executes one or more methods in the above-mentioned embodiments.

[0152] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required for the present application.

[0153] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0154] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some service interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0155] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0156] 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.

[0157] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory 305 and includes several instructions for a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory 305 includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk.

[0158] The above is only an exemplary embodiment of the present disclosure and cannot be used to limit the scope of the present disclosure. That is, any equivalent changes and modifications made according to the teachings of the present disclosure are still within the scope of the present disclosure. After considering the disclosure of the specification, those skilled in the art will easily think of other embodiments of the present disclosure. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary technical means in the technical field that are not recorded in the present disclosure. The description and examples are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for analyzing the performance of network element boards in the entire network, characterized in that: Applied to the whole network element single board performance analysis platform, the method includes: Divide the entities in the network structure according to the preset classification standard, and construct a tree-structured data model according to the division result, wherein the entities include a computer room, a switch and a network element board; Collect performance indicator data of all network element boards in real time and classify the performance indicator data into corresponding tree structure nodes; Dynamically generate a current threshold value according to current data in the network structure using a preset prediction model, and determine whether the performance indicator data of the network element board is normal according to the current threshold value; When the performance indicator data of a network element board is abnormal, a target network element board is determined, and the abnormality is located according to the path of the target network element board in the tree structure. The dividing of entities in the network structure according to the preset classification standard includes: The top layer of the network structure is used as a root node, and the network under the root node is divided according to the geographical location, so that each computer room or data center serves as an independent first-level node; Divide the devices under the first-level node according to function or device type, so that each device serves as an independent second-level node; The network element boards under the secondary node are divided according to the specific components of the device, so that each network element board acts as an independent third-level node. The classifying the performance indicator data into corresponding tree structure nodes includes: Establishing a first mapping rule between performance indicator data and third-level nodes, establishing a second mapping rule between performance indicator data and second-level nodes, and establishing a third mapping rule between performance indicator data and first-level nodes; Performing a first classification on the performance indicator data according to the first mapping rule; When there is first remaining data in the performance indicator data after the first classification is completed, performing a second classification on the first remaining data according to the second mapping rule, and performing a third classification on the first target data after the second classification is performed according to the tree structure; When the first remaining data has second remaining data after the second classification is completed, the second remaining data is fourthly classified according to the third mapping rule, and the second target data that has been subjected to the fourth classification is fifthly classified according to the tree structure.

2. The method for analyzing the performance of network element boards in the entire network according to claim 1, characterized in that: The data model of the tree structure constructed according to the division result includes: Setting a unique identifier for each node and setting corresponding attribute information according to the type and function of the node, wherein the attribute information includes name, location, function and performance parameters; According to the hierarchical relationship of the network structure, a parent-child relationship is established for each node, and a data model of a tree structure is constructed according to the parent-child relationship.

3. The method for analyzing the performance of network element boards in the entire network according to claim 1, characterized in that: The dynamically generating a current threshold value according to the current data in the network structure by using a preset prediction model comprises: Collecting first historical performance data from the network structure, and marking the first historical performance data as normal or abnormal according to the state of the network structure to form a training sample library; Using a support vector machine algorithm to train the training sample library to obtain a preliminary prediction model; Continuously collecting second historical performance data and evaluating the preliminary prediction model according to the second historical performance data, adjusting the parameters of the support vector machine algorithm according to the evaluation result to train the preset prediction model, wherein the generation time period of the second historical performance data is later than the generation time period of the first historical performance data; Current performance data is collected, and current thresholds of different types of performance data are generated based on the current performance data using the preset prediction model.

4. The method for analyzing the performance of network element boards in the entire network according to claim 1, characterized in that: The locating the abnormality according to the path of the target network element board in the tree structure comprises: Starting from the target network element board, searching for each target node level by level upward along the parent-child relationship of the tree structure; Obtaining a change trend of performance indicators between the target node and the child node of the target node, wherein the performance indicators include CPU usage, memory occupancy, and network throughput; Determine whether the target node is the root cause of the anomaly based on the change trend, the associated anomalies with other nodes, and the historical performance of the target node.

5. The method for analyzing the performance of network element boards in the entire network according to claim 1, characterized in that: The method further comprises: According to the flow distribution of the network structure, nodes with flow exceeding a preset threshold are taken as target nodes; On the target node, deploying one or more communication interfaces as redundant interfaces; The state of the primary interface of the target node is detected by a heartbeat packet, and when a fault signal is detected, the data channel is switched from the primary interface to a redundant interface for data transmission.

6. A system for analyzing the performance of network element boards in the entire network, characterized in that: It includes structure module, classification module, judgment module and positioning module, among which: A structure module is configured to divide entities in the network structure according to a preset classification standard and construct a tree-structured data model according to the division result, wherein the entities include a computer room, a switch and a network element board; A classification module is configured to collect performance indicator data of all network element boards in real time and classify the performance indicator data into corresponding tree structure nodes; A judgment module configured to dynamically generate a current threshold value according to current data in the network structure using a preset prediction model, and determine whether the performance indicator data of the network element board is normal according to the current threshold value; a positioning module configured to determine a target network element board when performance indicator data of a network element board is abnormal, and locate the abnormality according to a path of the target network element board in the tree structure, The dividing of entities in the network structure according to the preset classification standard includes: The top layer of the network structure is used as a root node, and the network under the root node is divided according to the geographical location, so that each computer room or data center serves as an independent first-level node; Divide the devices under the first-level node according to function or device type, so that each device serves as an independent second-level node; The network element boards under the secondary node are divided according to the specific components of the device, so that each network element board acts as an independent third-level node. The classifying the performance indicator data into corresponding tree structure nodes includes: Establishing a first mapping rule between performance indicator data and third-level nodes, establishing a second mapping rule between performance indicator data and second-level nodes, and establishing a third mapping rule between performance indicator data and first-level nodes; Performing a first classification on the performance indicator data according to the first mapping rule; When there is first remaining data in the performance indicator data after the first classification is completed, performing a second classification on the first remaining data according to the second mapping rule, and performing a third classification on the first target data after the second classification is performed according to the tree structure; When the first remaining data has second remaining data after the second classification is completed, the second remaining data is fourthly classified according to the third mapping rule, and the second target data that has been subjected to the fourth classification is fifthly classified according to the tree structure.

7. An electronic device, characterized in that: It includes a processor, a memory, a user interface and a network interface, the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory so that the electronic device executes the method as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 5 is performed.

Images