Cloud network fault detection method and device, equipment, storage medium and program product

Abstract

The invention relates to the technical field of cloud computing, and discloses a cloud network fault detection method, device and equipment, a storage medium and a program product. The method comprises the following steps: acquiring a topological relation, a connection relation and a flow trend of each node in a target cloud network, and constructing a cloud network topological graph; acquiring characteristic parameters and operation states of each node in the target cloud network, and constructing a network element data model in combination with the cloud network topological graph; obtaining a cloud network fault detection knowledge base, and constructing a network fault diagnosis rule graph in combination with the cloud network topological graph and the network element data model; and training based on the cloud network topological graph, the network element data model and the network fault diagnosis rule graph by adopting a graph neural network to obtain a cloud network fault detection model so as to carry out cloud network fault detection. According to the scheme, the efficiency is high and the accuracy is good when the cloud network fault detection function is realized.

Description

Technical Field

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

[0002] Cloud networks are based on cloud computing technology, integrating computing resources such as servers, storage devices, and databases, and realizing efficient resource utilization through centralized management and scheduling. As cloud networks become more popular, how to detect cloud network faults has become an urgent problem to be solved.

[0003] In the related technology, technical personnel with rich experience in problem solving and relevant professional knowledge reserves locate and diagnose faults in cloud networks. However, due to the high system complexity of cloud networks, there are many types of faults that occur, and the root causes of faults are difficult to locate. In addition, it is greatly affected by the subjective influence of technical personnel. The threshold for fault detection is high, the accuracy is low, and the efficiency is low. Summary of the invention

[0004] In view of this, an object of the present invention is to provide a cloud network fault detection method, apparatus, device, storage medium and program product to solve the problem of low efficiency when performing cloud network fault detection.

[0005] In a first aspect, the present invention provides a cloud network fault detection method, the method comprising:

[0006] Obtain the topological relationship, connection relationship and traffic direction of each node in the target cloud network, and build a cloud network topology map;

[0007] Acquire characteristic parameters and operation status of each node in the target cloud network, and build a network element data model in combination with the cloud network topology map;

[0008] Acquire a cloud network fault detection knowledge base, and construct a network fault diagnosis rule map by combining the cloud network topology map and the network element data model;

[0009] A graph neural network is adopted to perform training based on the cloud network topology map, the network element data model and the network fault diagnosis rule map to obtain a cloud network fault detection model for cloud network fault detection.

[0010] In an optional implementation, each node in the target cloud network includes a number of network element nodes and a number of host nodes;

[0011] The obtaining of the topological relationship, connection relationship and traffic direction of each node in the target cloud network and building a cloud network topology map includes:

[0012] Obtain the topological relationship, connection relationship, and traffic direction between each network element node and each host node in the target cloud network;

[0013] A cloud network topology map is constructed by taking each network element node and each host node as entities, the topological relationship and connection relationship between each network element node and each host node as entity attributes, and the traffic direction between each network element node and each host node as the edge between each entity.

[0014] In an optional implementation, the characteristic parameters of the node include matching paths and intermediate files for communication between the nodes, and the operation status of the node includes the connection status, working status and traffic distribution of each node;

[0015] The obtaining of characteristic parameters and operating status of each node in the target cloud network and building a network element data model in combination with the cloud network topology map includes:

[0016] Obtain the matching paths and intermediate files for communication between nodes in the target cloud network, as well as the connection status, working status and traffic distribution of each node, and build a network element data model based on the topological relationship, connection relationship and traffic direction of each node in the cloud network topology map.

[0017] In an optional implementation, the acquiring of a cloud network fault detection knowledge base, combining the cloud network topology map and the network element data model, and constructing a network fault diagnosis rule map includes:

[0018] Obtain cloud network expert experience, operation and maintenance documents, troubleshooting manuals, and historical fault records to build a cloud network fault problem detection knowledge base;

[0019] Based on the cloud network fault problem detection knowledge base, combined with the topological relationship, connection relationship and traffic direction of each node in the cloud network topology map and the characteristic parameters and operating status of each node in the network element data model, a cloud network fault diagnosis rule map is constructed with network fault problems as entities, judgment conditions of network fault problems as entity attributes, and associations between network fault problems as edges between entities.

[0020] In an optional implementation, a graph neural network is used to perform training based on the cloud network topology map, the network element data model, and the network fault diagnosis rule map to obtain a cloud network fault detection model, including:

[0021] The cloud network topology map, the network element data model and the network fault diagnosis rule map are used as data sets, and a graph neural network is used to learn the weights of the edges between the entities in the network fault diagnosis rule map to obtain a cloud network fault detection model.

[0022] In an optional implementation, the performing cloud network fault detection includes:

[0023] The cloud network fault detection model is used to collect characteristic parameters and operating status of each node in the network element data model;

[0024] Based on the network fault diagnosis rule map and the characteristic parameters and operating status of each node in the network element data model, cloud network fault inspection, positioning and reasoning are performed, and the root cause of the cloud network fault and the corresponding confidence probability are output.

[0025] In a second aspect, the present invention provides a cloud network fault detection device, the device comprising:

[0026] The topology map module is used to obtain the topological relationship, connection relationship and traffic direction of each node in the target cloud network and build a cloud network topology map;

[0027] A data model module is used to obtain characteristic parameters and operating status of each node in the target cloud network, and to construct a network element data model in combination with the cloud network topology map;

[0028] A rule map module is used to obtain a cloud network fault detection knowledge base, and to construct a network fault diagnosis rule map by combining the cloud network topology map and the network element data model;

[0029] The fault detection module is used to adopt a graph neural network to perform training based on the cloud network topology map, the network element data model and the network fault diagnosis rule map to obtain a cloud network fault detection model for cloud network fault detection.

[0030] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor are communicatively connected to each other, computer instructions are stored in the memory, and the processor executes the cloud network fault detection method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0031] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the cloud network fault detection method of the first aspect or any corresponding embodiment thereof.

[0032] In a fifth aspect, the present invention provides a computer program product, comprising computer instructions, wherein the computer instructions are used to enable a computer to execute the cloud network fault detection method of the above-mentioned first aspect or any corresponding embodiment thereof.

[0033] The technical solution provided by the present invention may include the following beneficial effects:

[0034] The cloud network fault detection method provided by the present invention first obtains the topological relationship, connection relationship and traffic direction of each node in the target cloud network, constructs a cloud network topology map, then obtains the characteristic parameters and operating status of each node in the target cloud network, combines the cloud network topology map, constructs a network element data model, then obtains a cloud network fault detection knowledge base, combines the cloud network topology map and the network element data model, constructs a network fault diagnosis rule map, and finally adopts a graph neural network to train based on the cloud network topology map, the network element data model and the network fault diagnosis rule map to obtain a cloud network fault detection model for cloud network fault detection. The above scheme, by combining the cloud network topology map, the network element data model and the cloud network fault detection knowledge base, can construct a more targeted network fault diagnosis rule map for the target cloud network, and train a more targeted cloud network fault detection model by combining the cloud network topology map, the network element data model and the network fault diagnosis rule map, without manual intervention, thereby improving the accuracy and efficiency of subsequent cloud network fault detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 is a flow chart of a cloud network fault detection method according to an embodiment of the present invention;

[0037] Figure 2 is a flow chart of another cloud network fault detection method according to an embodiment of the present invention;

[0038] Figure 3 is a schematic diagram of cloud network fault detection according to an embodiment of the present invention;

[0039] Figure 4 is a flow chart of cloud network fault detection according to an embodiment of the present invention;

[0040] Figure 5 is a structural block diagram of a cloud network fault detection device according to an embodiment of the present invention;

[0041] Figure 6 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0043] An embodiment of the present invention provides a cloud network fault detection method, which improves the efficiency of cloud network fault detection by combining a cloud network topology map, a network element data model, and a network fault diagnosis rule map to perform cloud network fault detection.

[0044] According to an embodiment of the present invention, an embodiment of a cloud network fault detection method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0045] In this embodiment, a cloud network fault detection method is provided, which can be used for laptop computers, desktop computers, servers, etc. Figure 1 is a flow chart of a cloud network fault detection method according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0046] Step S101, obtain the topological relationship, connection relationship and traffic direction of each node in the target cloud network, and build a cloud network topology map.

[0047] The target cloud network is the cloud network to be fault-detected. Each node in the target cloud network is each virtual device or physical device in the target cloud network. The topological relationship of each node can be star, ring, bus, tree, mesh, etc. The connection relationship of each node is whether each node is connected and the connection method, such as physical connection (such as network cable, optical fiber, etc.) or virtual connection (such as virtual network interface, etc.). The flow direction is the direction of flow transmission between each node. The cloud network topology map is constructed based on the knowledge graph. The knowledge graph is a structured knowledge representation method used to describe entities, concepts and their relationships in the real world. The knowledge graph stores and organizes knowledge in the form of a graph, where nodes represent entities or concepts, and edges represent the relationship between entities or concepts. Optionally, in the cloud network topology map, each node of the target cloud network can be treated as an entity, connected based on the topological relationship and connection relationship, and the direction of the edge is determined according to the flow direction.

[0048] Step S102, obtaining characteristic parameters and operating status of each node in the target cloud network, and building a network element model in combination with the cloud network topology map.

[0049] The characteristic parameters of each node in the target cloud network may be the address assigned to the node, how to communicate with other nodes, etc. The operating status of each node may be how many other nodes are currently connected, whether it is in a normal working state, how much traffic is allocated, etc. The network element data model is a structured data set used to describe and store the node characteristics, edge information, entity attributes and graph structure of each node in the target cloud network. In this embodiment, the network element data model can provide the topological relationship, connection relationship, traffic direction, characteristic parameters and operating status of each node in the target cloud network for subsequent cloud network fault detection.

[0050] Step S103, obtaining a cloud network fault detection knowledge base, combining the cloud network topology map and the network element data model, and constructing a network fault diagnosis rule map.

[0051] The cloud network fault detection knowledge base can be established in advance based on the collected network data, historical fault information, etc. The network fault diagnosis rule map is constructed based on the knowledge map, and various fault problems can be used as entities and the associations between various faults can be used as edges. The cloud network topology map and the network element data model can provide the network fault diagnosis rule map with various information about the target cloud network, so that the network fault diagnosis rule map not only contains content for various fault problems, but also content about the target cloud network, which can more specifically diagnose network faults on the target cloud network and improve the accuracy of the diagnosis results.

[0052] Step S104, using a graph neural network, training based on the cloud network topology map, the network element data model and the network fault diagnosis rule map to obtain a cloud network fault detection model for cloud network fault detection.

[0053] Graph neural network is a deep learning model that specializes in processing graph structure data. It can capture the dependencies between nodes and the topological structure of the graph. By using the cloud network topology map, the network element model, and the information in the network fault diagnosis rule map as data sets, the graph neural network is trained to obtain a cloud network fault detection model. The cloud network fault detection model can predict the connection between various fault problems of the target cloud network and trace the root cause of the fault, so that relevant technicians can perform maintenance based on the root cause of the fault, thereby improving the efficiency of cloud network fault detection.

[0054] The cloud network fault detection method provided in this embodiment first obtains the topological relationship, connection relationship and traffic direction of each node in the target cloud network, constructs a cloud network topology map, then obtains the characteristic parameters and operating status of each node in the target cloud network, combines the cloud network topology map, and constructs a network element data model, then obtains a cloud network fault detection knowledge base, combines the cloud network topology map and the network element data model, and constructs a network fault diagnosis rule map, and finally uses a graph neural network to train based on the cloud network topology map, the network element data model and the network fault diagnosis rule map to obtain a cloud network fault detection model for cloud network fault detection. The above scheme, by combining the cloud network topology map, the network element data model and the cloud network fault detection knowledge base, can construct a more targeted network fault diagnosis rule map for the target cloud network, and train a more targeted cloud network fault detection model by combining the cloud network topology map, the network element data model and the network fault diagnosis rule map, without manual intervention, thereby improving the accuracy and efficiency of subsequent cloud network fault detection.

[0055] In this embodiment, a cloud network fault detection method is provided, which can be used for laptop computers, desktop computers, servers, etc. Figure 2 is a flow chart of a cloud network fault detection method according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0056] Step S201, obtain the topological relationship, connection relationship and traffic direction of each node in the target cloud network, and build a cloud network topology map.

[0057] Specifically, the above step S201 includes:

[0058] Step S2011, obtaining the topological relationship, connection relationship and traffic direction between each network element node and each host node in the target cloud network.

[0059] Each node in the target cloud network includes several network element nodes and several host nodes. Network element refers to a network element, which is the basic unit that constitutes the target cloud network and is responsible for processing tasks such as data transmission, routing, and switching in the network. Exemplarily, the network element node can be a router, a switch, a load balancer, a firewall, a gateway, an SDN (software defined network) controller, etc., which can be a physical device or a virtual device. The host usually refers to a device that plays a leading role in the system, which can control and manage slave devices. The host itself is a computer system that can run independently. Exemplarily, the host node can be a physical machine or a virtual machine, and can be a terminal device, a client device, a server device, etc.

[0060] Step S2012, with each network element node and each host node as entities, the topological relationship and connection relationship between each network element node and each host node as entity attributes, and the traffic direction between each network element node and each host node as the edge between each entity, to build a cloud network topology map.

[0061] Among them, the weight of the edge between each node is determined based on the traffic between the nodes.

[0062] Optionally, when at least one of the topological relationship, connection relationship or traffic direction of each node in the target cloud network changes, the cloud network topology map is updated based on the changed topological relationship, connection relationship and traffic direction of each node.

[0063] Step S202, obtaining characteristic parameters and operating status of each node in the target cloud network, and building a network element model in combination with the cloud network topology map.

[0064] Optionally, the characteristic parameters of the node include the matching paths and intermediate files for communication between each node, and the running state of the node includes the connection state, working state and traffic distribution of each node. The intermediate file can be a routing table, a forwarding table, etc. between physical machines, or a flow table between virtual machines, or an intermediate file for communication between the cloud and the cloud. The characteristic parameters of the node can also include IP address, subnet mask, default gateway, server address, MAC address, bandwidth utilization, packet loss rate, delay, etc. The connection state of the node can be the number of established connections and the stability of the connection, the working state can be a normal working state up, unavailable or offline state down, and the traffic distribution can be the amount of inbound data, outbound data, data transmission rate, etc. flowing through the node.

[0065] Specifically, the matching paths and intermediate files for communication between nodes in the target cloud network, as well as the connection status, working status and traffic distribution of each node are obtained, and the network element data model is constructed by combining the topological relationship, connection relationship and traffic direction of each node in the cloud network topology map.

[0066] Optionally, when the characteristic parameters and operating status of each node in the cloud network topology map and / or the target cloud network change, the network element data model is updated based on the changed characteristic parameters and operating status of each node in the cloud network topology map and / or the target cloud network.

[0067] Step S203, obtaining a cloud network fault detection knowledge base, combining the cloud network topology map and the network element data model, and constructing a network fault diagnosis rule map.

[0068] Specifically, the above step S203 includes:

[0069] Step S2031, obtain cloud network expert experience, operation and maintenance documents, troubleshooting manuals and historical fault records to build a cloud network fault problem detection knowledge base.

[0070] The cloud network expert experience, operation and maintenance documents, troubleshooting manuals and historical fault records contain various network fault events, various network fault objects and the relationship between various network fault events and network fault objects in the cloud network. By performing semantic recognition on the cloud network expert experience, operation and maintenance documents, troubleshooting manuals and historical fault records, various network fault events, various network fault objects and the relationship between various network fault events and network fault objects can be extracted to build a cloud network fault problem detection knowledge base.

[0071] Step S2032, based on the cloud network fault problem detection knowledge base, combined with the topological relationship, connection relationship and traffic direction of each node in the cloud network topology map and the characteristic parameters and operating status of each node in the network element data model, with the network fault problem as the entity, the judgment condition of the network fault problem as the entity attribute, and the association relationship between the network fault problems as the edge between each entity, to construct a cloud network fault diagnosis rule map.

[0072] The relevant information of the target cloud network provided by the cloud network topology map and the network element data model is mapped to the cloud network fault problem detection process through labels, and the data in the cloud network fault problem detection knowledge base is called through the corresponding interface to perform entity recognition and relationship extraction on the data in the cloud network fault problem detection knowledge base, thereby constructing a cloud network fault diagnosis rule map. Network fault problems include network fault events and network fault objects. The judgment conditions of network fault problems can be judgment rules, judgment parameters, judgment states, etc. For example, when the parameter is 0, it is judged to be faulty or when it is offline, it is judged to be faulty. The edge between two entities represents the association relationship between the two entities, and the weight of the edge represents the degree of association between the two entities. The triples of the cloud network fault diagnosis rule map indicate the logical relationship and degree of association between the network fault problems corresponding to each entity.

[0073] Optionally, when the cloud network fault problem detection knowledge base, cloud network topology map and network element data model change, the cloud network fault diagnosis rule map is updated based on the changed cloud network fault problem detection knowledge base, cloud network topology map and network element data model.

[0074] Step S204, using a graph neural network, training based on the cloud network topology map, the network element data model and the network fault diagnosis rule map to obtain a cloud network fault detection model for cloud network fault detection.

[0075] The entities in the network fault diagnosis rule map represent different types of network fault problems, while the nodes in the cloud network topology map represent network elements or hosts. By constructing an association matrix between network fault problems and root causes of faults, the relevant information of the target cloud network in the cloud network topology map can be linked to the network fault problems and root causes of faults in the network fault diagnosis rule map, thereby realizing reasoning from the network fault problems of the target cloud network to the root causes of faults. The characteristic parameters and operating status of the nodes in the target cloud network in the network element metadata model can provide the graph neural network with information such as node features, edge information, entity attributes, graph structure, adjacency matrix, etc., so that the graph neural network can learn the embedded representation of nodes, capture the complex relationship between nodes, and realize the cloud network fault detection function. The multi-task learning framework can also be used to simultaneously process the cloud network topology map, the network element metadata model, and the network fault diagnosis rule map, share the features in the cloud network topology map, the network element metadata model, and the network fault diagnosis rule map when training the graph neural network, optimize the parameters of the graph neural network, improve the accuracy of the obtained cloud network fault detection model, and thus improve the accuracy of subsequent cloud network fault detection through the cloud network fault detection model.

[0076] Specifically, the cloud network topology map, the network element data model and the network fault diagnosis rule map are used as data sets, and a graph neural network is used to learn the weights of the edges between the entities in the network fault diagnosis rule map to obtain a cloud network fault detection model.

[0077] After obtaining the cloud network fault detection model, the target cloud network can be detected (online reasoning) through the cloud network fault detection model. The detection time interval can be set for fault detection, such as once a day. Fault detection can also be actively triggered through corresponding instructions or functions. When performing fault detection, the cloud network fault detection model is first used to collect the characteristic parameters and operating status of each node in the network element model. Then, based on the network fault diagnosis rule map and the characteristic parameters and operating status of each node in the network element model, the cloud network fault inspection is performed. If a fault is detected in the target cloud network, the detected fault is located and inferred, and the root cause of the cloud network fault and the corresponding confidence probability are output. Multiple root causes of cloud network faults and the corresponding confidence probabilities can be output at one time for reference by relevant technical personnel and timely maintenance of the target cloud network. Relevant technical personnel can also fine-tune and update the cloud network fault detection model according to their needs, such as updating when new improvements are found during the fault detection process.

[0078] The cloud network fault detection method provided in this embodiment first obtains the topological relationship, connection relationship and traffic direction of each node in the target cloud network, constructs a cloud network topology map, then obtains the characteristic parameters and operating status of each node in the target cloud network, combines the cloud network topology map, and constructs a network element data model, then obtains a cloud network fault detection knowledge base, combines the cloud network topology map and the network element data model, and constructs a network fault diagnosis rule map, and finally uses a graph neural network to train based on the cloud network topology map, the network element data model and the network fault diagnosis rule map to obtain a cloud network fault detection model for cloud network fault detection. The above scheme, by combining the cloud network topology map, the network element data model and the cloud network fault detection knowledge base, can construct a more targeted network fault diagnosis rule map for the target cloud network, and train a more targeted cloud network fault detection model by combining the cloud network topology map, the network element data model and the network fault diagnosis rule map, without manual intervention, thereby improving the accuracy and efficiency of subsequent cloud network fault detection.

[0079] As one or more specific application embodiments of the embodiments of the present invention, the optimal implementation scheme or the scheme that the inventor most wants to embody is described below in combination with specific application scenarios.

[0080] Figure 3 is a schematic diagram of cloud network fault detection according to an embodiment of the present invention, Figure 4 is a flow chart of cloud network fault detection according to an embodiment of the present invention. Figure 3 As shown in the figure, data collection is first performed to build a network topology map and network element data model. Then, the network topology map and network element data model are combined with the network problem diagnosis knowledge base to perform entity recognition and relationship extraction, and to build a network problem diagnosis rule map. Figure 4 As shown in the figure, data collection and knowledge extraction are first performed, and then entity recognition and relationship extraction are performed to construct a network topology map and a network problem diagnosis rule map. Then, the graph neural network is trained, and then online reasoning is performed through the trained graph neural network (i.e., the cloud network fault detection model), and the knowledge base is updated in a timely manner.

[0081] In this embodiment, a cloud network fault detection device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0082] This embodiment provides a cloud network fault detection device, such as Figure 5 As shown, including:

[0083] The topology map module 501 is used to obtain the topological relationship, connection relationship and traffic direction of each node in the target cloud network and construct a cloud network topology map;

[0084] The data model module 502 is used to obtain characteristic parameters and operation status of each node in the target cloud network, and to construct a network element data model in combination with the cloud network topology map;

[0085] The rule map module 503 is used to obtain a cloud network fault detection knowledge base, and to construct a network fault diagnosis rule map in combination with the cloud network topology map and the network element data model;

[0086] The fault detection module 504 is used to adopt a graph neural network to perform training based on the cloud network topology map, the network element data model and the network fault diagnosis rule map to obtain a cloud network fault detection model for cloud network fault detection.

[0087] In an optional embodiment, each node in the target cloud network includes a number of network element nodes and a number of host nodes; the topology map module is also used to: obtain the topological relationship, connection relationship and traffic direction between each network element node and each host node in the target cloud network; take each network element node and each host node as entities, the topological relationship and connection relationship between each network element node and each host node as entity attributes, and take the traffic direction between each network element node and each host node as the edge between each entity to construct a cloud network topology map.

[0088] In an optional implementation, the characteristic parameters of the node include matching paths and intermediate files for communication between nodes, and the operating status of the node includes the connection status, working status and traffic distribution of each node; the data model module is also used to: obtain the matching paths and intermediate files for communication between nodes in the target cloud network, as well as the connection status, working status and traffic distribution of each node, and construct a network element data model in combination with the topological relationship, connection relationship and traffic direction of each node in the cloud network topology map.

[0089] In an optional implementation, the rule graph module is also used to: obtain cloud network expert experience, operation and maintenance documents, troubleshooting manuals and historical fault records to build a cloud network fault problem detection knowledge base; based on the cloud network fault problem detection knowledge base, combined with the topological relationship, connection relationship and traffic direction of each node in the cloud network topology map and the characteristic parameters and operating status of each node in the network element data model, with the network fault problem as the entity, the judgment condition of the network fault problem as the entity attribute, and the association relationship between the network fault problems as the edge between each entity, to build a cloud network fault diagnosis rule graph.

[0090] In an optional embodiment, the fault detection module is also used to: use the cloud network topology map, the network element data model and the network fault diagnosis rule map as data sets, use a graph neural network to learn the weights of the edges between each entity in the network fault diagnosis rule map, and obtain a cloud network fault detection model.

[0091] In an optional embodiment, the fault detection module is also used to: use the cloud network fault detection model to collect the characteristic parameters and operating status of each node in the network element data model; based on the network fault diagnosis rule map and the characteristic parameters and operating status of each node in the network element data model, perform cloud network fault inspection, location and reasoning, and output the root cause of the cloud network fault and the corresponding confidence probability.

[0092] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0093] The cloud network fault detection device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0094] The embodiment of the present invention also provides a computer device having the above Figure 5 The cloud network fault detection device is shown.

[0095] See also Figure 6 , Figure 6 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 6 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 6 A processor 10 is taken as an example.

[0096] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0097] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0098] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0099] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0100] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Figure 6 The example of connecting through bus is taken in the following.

[0101] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device (e.g., an LED) and a tactile feedback device (e.g., a vibration motor), etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0102] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0103] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0104] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations shall all fall within the scope of protection of the present invention.

Claims

1. A cloud network fault detection method, characterized in that: The method comprises: Obtain the topological relationship, connection relationship and traffic direction of each node in the target cloud network, and build a cloud network topology map; Acquire characteristic parameters and operation status of each node in the target cloud network, and build a network element data model in combination with the cloud network topology map; Acquire a cloud network fault detection knowledge base, and construct a network fault diagnosis rule map by combining the cloud network topology map and the network element data model; A graph neural network is adopted to perform training based on the cloud network topology map, the network element data model and the network fault diagnosis rule map to obtain a cloud network fault detection model for cloud network fault detection.

2. The method according to claim 1, characterized in that Each node in the target cloud network includes a number of network element nodes and a number of host nodes; The obtaining of the topological relationship, connection relationship and traffic direction of each node in the target cloud network and building a cloud network topology map includes: Obtain the topological relationship, connection relationship, and traffic direction between each network element node and each host node in the target cloud network; A cloud network topology map is constructed by taking each network element node and each host node as entities, the topological relationship and connection relationship between each network element node and each host node as entity attributes, and the traffic direction between each network element node and each host node as the edge between each entity.

3. The method according to claim 2, characterized in that The characteristic parameters of the nodes include the matching paths and intermediate files for communication between the nodes, and the operation status of the nodes includes the connection status, working status and traffic distribution of the nodes; The obtaining of characteristic parameters and operating status of each node in the target cloud network and building a network element data model in combination with the cloud network topology map includes: Obtain the matching paths and intermediate files for communication between nodes in the target cloud network, as well as the connection status, working status and traffic distribution of each node, and build a network element data model based on the topological relationship, connection relationship and traffic direction of each node in the cloud network topology map.

4. The method according to claim 3, characterized in that The acquiring of the cloud network fault detection knowledge base, combining the cloud network topology map and the network element data model, and constructing a network fault diagnosis rule map includes: Obtain cloud network expert experience, operation and maintenance documents, troubleshooting manuals, and historical fault records to build a cloud network fault problem detection knowledge base; Based on the cloud network fault problem detection knowledge base, combined with the topological relationship, connection relationship and traffic direction of each node in the cloud network topology map and the characteristic parameters and operating status of each node in the network element data model, a cloud network fault diagnosis rule map is constructed with network fault problems as entities, judgment conditions of network fault problems as entity attributes, and associations between network fault problems as edges between entities.

5. The method according to any one of claims 1 to 4, characterized in that: A graph neural network is adopted to perform training based on the cloud network topology map, the network element data model and the network fault diagnosis rule map to obtain a cloud network fault detection model, including: The cloud network topology map, the network element data model and the network fault diagnosis rule map are used as data sets, and a graph neural network is used to learn the weights of the edges between the entities in the network fault diagnosis rule map to obtain a cloud network fault detection model.

6. The method according to claim 5, characterized in that The cloud network fault detection includes: The cloud network fault detection model is used to collect characteristic parameters and operating status of each node in the network element data model; Based on the network fault diagnosis rule map and the characteristic parameters and operating status of each node in the network element data model, cloud network fault inspection, positioning and reasoning are performed, and the root cause of the cloud network fault and the corresponding confidence probability are output.

7. A cloud network fault detection device, characterized in that: The device comprises: The topology map module is used to obtain the topological relationship, connection relationship and traffic direction of each node in the target cloud network and build a cloud network topology map; A data model module is used to obtain characteristic parameters and operating status of each node in the target cloud network, and to construct a network element data model in combination with the cloud network topology map; A rule map module is used to obtain a cloud network fault detection knowledge base, and to construct a network fault diagnosis rule map by combining the cloud network topology map and the network element data model; The fault detection module is used to adopt a graph neural network to perform training based on the cloud network topology map, the network element data model and the network fault diagnosis rule map to obtain a cloud network fault detection model for cloud network fault detection.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the cloud network fault detection method according to any one of claims 1 to 6 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the cloud network fault detection method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The method comprises computer instructions for causing a computer to execute the cloud network fault detection method according to any one of claims 1 to 6.

Images