CN114785674A - Fault positioning method and device, and computer-storable medium - Google Patents

Fault positioning method and device, and computer-storable medium Download PDF

Info

Publication number
CN114785674A
CN114785674A CN202210453803.7A CN202210453803A CN114785674A CN 114785674 A CN114785674 A CN 114785674A CN 202210453803 A CN202210453803 A CN 202210453803A CN 114785674 A CN114785674 A CN 114785674A
Authority
CN
China
Prior art keywords
entities
alarm
fault
graph
root
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210453803.7A
Other languages
Chinese (zh)
Inventor
梅承力
齐文
夏旭
邢燕霞
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Telecom Corp Ltd
Original Assignee
China Telecom Corp Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Telecom Corp Ltd filed Critical China Telecom Corp Ltd
Priority to CN202210453803.7A priority Critical patent/CN114785674A/en
Publication of CN114785674A publication Critical patent/CN114785674A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/06—Management of faults, events, alarms or notifications
    • H04L41/0677—Localisation of faults
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/06—Management of faults, events, alarms or notifications
    • H04L41/0631—Management of faults, events, alarms or notifications using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/06—Management of faults, events, alarms or notifications
    • H04L41/0631—Management of faults, events, alarms or notifications using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis
    • H04L41/065—Management of faults, events, alarms or notifications using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis involving logical or physical relationship, e.g. grouping and hierarchies
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/04—Arrangements for maintaining operational condition

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Databases & Information Systems (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The disclosure relates to a fault positioning method and device and a computer-readable storage medium, and relates to the technical field of networks. The fault positioning method comprises the steps of obtaining an alarm knowledge graph, wherein the alarm knowledge graph comprises entities and relations among the entities, and the entities comprise fault entities, root alarm entities generated by the fault entities and derivative alarm entities derived from the root alarm entities; carrying out relation reasoning on the knowledge graph by using a graph neural network; and determining the currently generated root alarm according to the result of the relationship inference and the current alarm data by using the classifier so as to position the fault. According to this disclosure, the fault location efficiency is improved.

Description

Fault positioning method and device, and computer-storable medium
Technical Field
The present disclosure relates to the field of network technologies, and in particular, to a fault location method and apparatus, and a computer-readable storage medium.
Background
The current 5G network adopts technologies such as clouding, virtualization, slicing and the like. Compared with the communication technology before 5G, the technology such as virtualization of the 5G mobile network causes complexity increase of a network level and a network architecture, fault points of a network slice increase, and root alarms and derived alarms caused by the fault points also increase. In addition, the coming of the 6G era can further increase the complexity of the network, and the types of fault points, root alarms and derived alarms can also increase.
In the related art, most of the existing network virtualization and slicing knowledge bases are represented by structured and unstructured text information, and when a network slicing fault occurs, the fault needs to be checked one by one according to the text information, for example, manual fault clearing and reasoning are adopted.
Disclosure of Invention
According to a first aspect of the present disclosure, there is provided a fault location method, including: acquiring an alarm knowledge graph, wherein the alarm knowledge graph comprises entities and relations among the entities, and the entities comprise fault entities, root alarm entities generated by the fault entities and derivative alarm entities derived from the root alarm entities; carrying out relation reasoning on the knowledge graph by using a graph neural network; and determining the currently generated root alarm according to the result of the relationship inference and the current alarm data by using the classifier so as to position the fault.
In some embodiments, the performing relational inference on the knowledge graph using the graph neural network includes:
acquiring a characteristic vector of an entity in a knowledge graph as an initial vector of a node in a graph neural network;
and carrying out relationship reasoning according to the initial vector of the node in the graph neural network to obtain a final vector of the node, wherein the final vector of the node corresponds to the root alarm.
In some embodiments, the performing a relationship inference according to the initial vectors of the nodes in the graph neural network to obtain a final vector of the nodes, as a result of the relationship inference, includes:
and determining the feature vector of the node at the next layer of the graph neural network according to the feature vector of the node at the current layer of the graph neural network and the feature vectors of other nodes having relations with the node at the current layer of the graph neural network until determining the feature vector of the node at the last layer of the graph neural network as the final vector of the node.
In some embodiments, the determining, by the classifier, a currently occurring root alarm according to the result of the relationship inference and the current alarm data to locate the fault includes:
determining the current occurrence probability of one or more root alarms according to the result of the relational reasoning and the current alarm data;
and taking the root alarm with the maximum probability as the currently occurring root alarm.
In some embodiments, the determining, by the classifier, a currently occurring root alarm according to the result of the relationship inference and the current alarm data to locate the fault includes:
and positioning the fault according to the current root alarm and the fault position information in the current alarm data.
In some embodiments, the fault location method further includes:
determining entities and the relation between the entities according to historical alarm data;
and establishing an alarm knowledge graph according to the entities and the relationship between the entities.
In some embodiments, the determining the entities and the relationships between the entities according to the historical alarm data includes:
and determining the entities and the relationship among the entities according to the structured data in the historical alarm data.
In some embodiments, the determining the entities and the relationships between the entities according to the historical alarm data includes:
and carrying out knowledge extraction on the semi-structured and unstructured historical alarm data to obtain entities and the relation among the entities.
In some embodiments, the establishing an alarm knowledge graph according to alarm knowledge includes:
carrying out knowledge fusion on the entities and the relation between the entities;
and establishing an alarm knowledge graph according to the result of knowledge fusion.
In some embodiments, the fault location method further comprises:
and reporting the positioned fault to a sub-slice management function, and processing the fault by the sub-slice management function.
According to a second aspect of the present disclosure, there is provided a fault location device comprising
The system comprises an acquisition module, a processing module and a display module, wherein the acquisition module is configured to acquire an alarm knowledge graph, the alarm knowledge graph comprises entities and relations among the entities, and the entities comprise fault entities, root alarm entities generated by the fault entities and derivative alarm entities derived from the root alarm entities;
an inference module configured to perform relational inference on the knowledge graph using the graph neural network;
and the positioning module is configured to determine a currently generated root alarm according to the result of the relation inference and the current alarm data by utilizing the classifier so as to position the fault.
According to a third aspect of the present disclosure, there is provided a fault location device comprising:
a memory; and
a processor coupled to the memory, the processor configured to perform the fault location method of any embodiment of the present disclosure based on instructions stored in the memory.
According to a fourth aspect of the present disclosure, there is provided a network management system comprising:
the fault locating device according to any one of the embodiments of the present disclosure.
According to a fifth aspect of the present disclosure, there is provided a computer-readable storage medium, on which a computer program is stored, which program, when executed by a processor, implements a fault localization method according to any one of the embodiments of the present disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description, serve to explain the principles of the disclosure.
The present disclosure may be more clearly understood from the following detailed description taken in conjunction with the accompanying drawings, in which:
FIG. 1 illustrates a flow diagram of a fault location method in accordance with some embodiments of the present disclosure;
FIG. 2 illustrates a schematic diagram of an alarm knowledge-graph in accordance with some embodiments of the present disclosure;
FIG. 3 illustrates a schematic diagram of knowledge inference according to some embodiments of the present disclosure;
FIG. 4 illustrates a schematic diagram of determining a root alarm in accordance with some embodiments of the present disclosure;
FIG. 5 illustrates a schematic diagram of constructing an alarm knowledge graph, according to some embodiments of the present disclosure;
FIG. 6 illustrates a schematic diagram of a fault location method according to some embodiments of the present disclosure;
FIG. 7 illustrates a schematic diagram of a method of fault handling according to some embodiments of the present disclosure;
FIG. 8 illustrates a block diagram of a fault locating device according to some embodiments of the present disclosure;
FIG. 9 shows a block diagram of a fault locating device according to further embodiments of the present disclosure;
figure 10 shows a schematic diagram of a network management system according to some embodiments of the present disclosure;
FIG. 11 illustrates a block diagram of a computer system for implementing some embodiments of the present disclosure.
Detailed Description
Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that: the relative arrangement of the components and steps, the numerical expressions, and numerical values set forth in these embodiments do not limit the scope of the present disclosure unless specifically stated otherwise.
Meanwhile, it should be understood that the sizes of the respective portions shown in the drawings are not drawn in an actual proportional relationship for the convenience of description.
The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the disclosure, its application, or uses.
Techniques, methods, and apparatus known to one of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate.
In all examples shown and discussed herein, any particular value should be construed as exemplary only and not as limiting. Thus, other examples of the exemplary embodiments may have different values.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, further discussion thereof is not required in subsequent figures.
In the related art, the method for checking the possible faults one by one has low positioning efficiency and inaccurate positioning result in a complex scene. Moreover, the result of manually positioning the fault is inaccurate, and the problem of how to position the network fault in a complex scene is difficult to solve.
In order to solve the above problems, the present disclosure provides a fault location method and apparatus, and a computer-readable storage medium.
Fig. 1 illustrates a flow diagram of a fault location method in accordance with some embodiments of the present disclosure.
As shown in fig. 1, the fault location method includes steps S1 to S3. In some embodiments, the following fault localization method is performed by a fault localization apparatus.
In step S1, an alarm knowledge graph is obtained, where the alarm knowledge graph includes entities and relationships between the entities, and the entities include a fault entity, a root alarm entity generated by the fault entity, and a derivative alarm entity derived from the root alarm entity.
The root alarm is used to indicate the type of failure that occurred, such as physical resource failure and virtual resource failure. The physical resource failure includes a failure of a device, a failure of a link, and the like. More specifically, the failure of the device may be, for example, a chip failure of the device, a device interface failure, or the like. Virtual resource failures include failures of NFVO (Network Function virtualization orchestrator), VNFM (virtualized Network Function Manager), VIM (virtualization in infrastructure Manager), and the like.
The fault entity indicates a more specific fault and a fault location, for example, a chip fault occurs in the a device of the machine room No. 1.
The derived alarm entity refers to a derived alarm generated by the root alarm. For example, the device status light is on, the device temperature is too high, and so on.
FIG. 2 illustrates a schematic diagram of an alarm knowledge-graph according to some embodiments of the present disclosure.
As shown in fig. 2, fault 1 may generate root alarm 2. Thus, in the knowledge-graph, there is a "yield" relationship between fault 1 and root alarm 2. Similarly, there is a "derivative" relationship between root alarm 1 and derivative alarm 1.
In step S2, the relation of the knowledge graph is inferred using the graph neural network.
In some embodiments, the using the graph neural network to perform the relationship inference on the knowledge graph includes: acquiring a characteristic vector of an entity in a knowledge graph as an initial vector of a node in a graph neural network; and carrying out relationship reasoning according to the initial vector of the node in the graph neural network to obtain a final vector of the node, wherein the final vector of the node corresponds to the root alarm.
FIG. 3 illustrates a schematic diagram of knowledge inference according to some embodiments of the present disclosure.
As shown in fig. 3, the input of the graph neural network is an alarm knowledge graph or a sub-graph of the alarm knowledge graph, which is represented as g ═ V, a }, where V is an entity set of the knowledge graph or the sub-graph thereof, and a represents a connection matrix of the graph structure, i.e., a connection relationship between entities. In fig. 3, 1 denotes a faulty entity, 2 and 4 are root alarms, and 3 is a derived alarm.
Before the alarm knowledge graph is input into the graph neural network, the entity in the knowledge graph is represented vectorially by using a network representation learning method, so that knowledge such as the relation between the role and the entity in the complex knowledge graph is changed into structural characteristics, and the graph neural network can learn the knowledge in the knowledge graph. The feature vector of each entity is then used as the initial vector of nodes in the graph neural network, i.e., the input to the graph neural network.
In some embodiments, the performing a relationship inference according to the initial vectors of the nodes in the graph neural network to obtain a final vector of the nodes, as a result of the relationship inference, includes: and determining the feature vector of the node at the next layer of the graph neural network according to the feature vector of the node at the current layer of the graph neural network and the feature vectors of other nodes having relations with the node at the current layer of the graph neural network until determining the feature vector of the node at the last layer of the graph neural network as the final vector of the node.
The graph neural network is a connection model, dependency relationships in a graph are obtained through information transfer between nodes in the network, the graph neural network updates the state of the nodes through neighbors at any depth from the nodes, the state can represent state information, and the graph neural network can obtain inference relationships between entities. The graph neural network may include a plurality of hidden layers. As shown in FIG. 3, Hn t-1，Hn tThe method is characterized in that the nth node is represented by hidden layers at T-1 and T, the graph neural network updates the representation of the nth node at T according to the nth node and the representation of the adjacent nodes (1 st and 3 rd nodes) which are connected with the nth node at T-1, and so on, and after updating T (T is a positive integer) times, the final vector (o) of the node n is obtainedn). The final vector of the node n corresponds to the vector of one root alarm, and the relationship inference refers to the dependency relationship inference from the node n to the root alarm.
The output of the graph neural network (i.e., the result of the relationship inference) is the final state of all nodes, denoted O (O)1,o2……,ov)。
In step S3, a classifier is used to determine the root alarm currently occurring according to the result of the relational inference and the current alarm data, so as to locate the fault.
In some embodiments, determining, using the classifier, a currently occurring root alarm to locate the fault based on the result of the relational inference and the current alarm data comprises: determining the current occurrence probability of one or more root alarms according to the result of the relationship inference and the current alarm data; and taking the root alarm with the maximum probability as the currently occurring root alarm.
FIG. 4 illustrates a schematic diagram for determining a root alarm in accordance with some embodiments of the present disclosure.
As shown in fig. 4, after the knowledgegraph + graph Neural Network, a simple classifier, such as FCNN, (Full volume Neural Network) is deployed. The current alarm data may include current derived alarm information, and the derived alarms and the root alarms are not in a one-to-one correspondence relationship, and in the knowledge graph, although the root alarm having a derived relationship with the current derived alarm may be found, a plurality of root alarms may be found, so that it cannot be determined by the knowledge graph only which type of root alarm currently occurs. The purpose of the knowledge-graph + graph neural network + classifier is to determine which root alarm caused the current derived alarm.
And calculating the current occurrence probability of each root alarm by utilizing the classifier and combining the current alarm data with the result output by the graph neural network. For example, the probability of the current alarm data being generated by root alarm 1 is 90%. The root alarm, of which the probability is the greatest, may then be taken as the currently occurring root alarm.
In some embodiments, determining, by the classifier, a currently occurring root alarm according to the result of the relational inference and the current alarm data to locate the fault includes: and positioning the fault according to the current root alarm and the fault position information in the current alarm data.
The current alarm data may comprise time information, location information, alarm level, priority, etc. in addition to the derived alarm data.
For example, the classifier has already calculated R _ LOS (LOSs of Signal) as the currently occurring root alarm, and then according to the alarm location information LOC1Can be located at LOC1The signal receiving apparatus B at the location malfunctions.
The present disclosure automatically infers relationships between alarms based on knowledge-graphs and graph-neural networks. When a fault occurs, the established reasoning relation can be utilized to quickly determine the root alarm, so that the fault is positioned, the fault troubleshooting time is shortened, and the influence of the fault on the service is reduced.
In some embodiments, the fault location method further comprises determining entities and relationships between entities based on historical alarm data; and establishing an alarm knowledge graph according to the entities and the relationship among the entities.
In some embodiments, historical alarm data is obtained from an alarm database. Historical alarm data can be divided into structured data, semi-structured data, and unstructured data. The alarm data includes an alarm type, an alarm ID, time information, location information, an alarm level, a priority, a relationship between an alarm and an alarm, and the like.
In some embodiments, determining the entities and relationships between the entities based on historical alarm data includes preprocessing the alarm data. The preprocessing of the alarm data is mainly the cleaning of the alarm data, such as time synchronization processing, missing and redundant data processing.
And after the data preprocessing is finished, constructing an alarm knowledge graph according to the cleaned data.
FIG. 5 illustrates a schematic diagram of constructing an alarm knowledge graph according to some embodiments of the present disclosure.
First, the entities and relationships in the alarm knowledge graph need to be determined. As shown in fig. 5, the structured data and the semi-structured, unstructured data are processed differently.
In some embodiments, determining entities and relationships between entities based on historical alarm data includes: and determining the entities and the relationship among the entities according to the structured data in the historical alarm data.
In some embodiments, determining entities, and relationships between entities, based on historical alarm data includes: and carrying out knowledge extraction on the semi-structured and unstructured historical alarm data to obtain entities and the relation among the entities. The knowledge extraction comprises entity extraction, relation extraction and attribute extraction. The attributes of the entity include an alarm ID, alarm time information, location information, alarm level, priority, etc.
In some embodiments, based on the alarm knowledge, an alarm knowledge graph is established, comprising: carrying out knowledge fusion on the entities and the relation between the entities; and establishing an alarm knowledge graph according to the result of knowledge fusion.
As shown in fig. 5, the knowledge fusion includes data fusion, knowledge inference, knowledge update, and the like. Through knowledge fusion, ambiguity between an entity or relationship and a corresponding fact object can be eliminated, and high-quality knowledge is formed.
And finally, expressing the knowledge after the knowledge fusion as a knowledge graph of a graph structure, specifically, expressing the entities by using nodes, expressing the relationship among the entities by using edges, and expressing the relationship among the fault entity, the root alarm and the derived alarm by using an entity-relationship-entity triple.
The method and the device represent the alarm data in the text form into the knowledge map form, and can utilize the knowledge map to intuitively infer the relation between alarms, so that the efficiency of fault positioning is improved.
Fig. 6 illustrates a schematic diagram of a fault location method according to some embodiments of the present disclosure.
As shown in FIG. 6, first, historical alarm data is obtained from the alarm database, and then the data is preprocessed. And carrying out operations such as knowledge extraction on the preprocessed data, and constructing an alarm knowledge graph. And then, taking the alarm knowledge graph as input, and carrying out rule reasoning on the relationship by using a graph neural network to generate a reasoning result. When alarm occurs, the classifier is used for finding out root alarm according to the current alarm data and the inference result. And finally, finding out the current fault according to the root alarm.
In some embodiments, the fault location method further comprises reporting the located fault to a sub-slice management function, the fault being handled by the sub-slice management function.
Fig. 7 illustrates a signaling diagram of a method of fault handling according to some embodiments of the present disclosure.
As shown in FIG. 7, at event 710, the NMS (Network Management System) locates the faulty entity.
In event 720, the NMS reports the failure to the NSSMF (Network Slice sub-Network Management Function).
In event 730, NSSMF handles traffic and network failures for the slice. For example, NSSMF may instruct VNF to isolate or shut down for a failure, add a new network function to take over the failed network function; or modify existing network functions to compensate for the failed part.
In event 740, the NSSMF notifies the NFVO to adjust the lifecycle and resources of the corresponding NS (Network Services) and VNF (virtual Network Function).
At event 750, NFVO performs NS failure handling and network-related physical resource, virtual resource failure handling based on the policy.
At event 760, the NFVM notifies the VIM to adjust NFVI (NFV Infrastructure).
At event 770, the VIM notifies the NFVI to make resource adjustments.
According to the method and the device, after the fault is located, the fault can be quickly isolated, and recovery measures are taken, so that the influence of the fault on service is reduced to the maximum extent. Once a fault occurs, new functional entities and physical resources can be rapidly assigned through NSSMF, services are switched, and smooth service is guaranteed.
Fig. 8 illustrates a block diagram of a fault location device, according to some embodiments of the present disclosure.
As shown in fig. 8, the fault locating device 80 includes an acquisition module 801.
An obtaining module 801 configured to obtain an alarm knowledge graph, where the alarm knowledge graph includes entities and relationships between the entities, and the entities include a fault entity, a root alarm entity generated by the fault entity, and a derivative alarm entity derived from the root alarm entity, for example, execute step S1 shown in fig. 1.
The inference module 802, configured to utilize the graph neural network, performs a relational inference on the knowledge graph, for example, performing step S2 as shown in fig. 1.
And a positioning module 803 configured to determine a currently occurring root alarm according to the result of the relationship inference and the current alarm data by using the classifier to position the fault. For example, step S3 shown in fig. 1 is performed.
FIG. 9 shows a block diagram of a fault location device, according to further embodiments of the present disclosure.
As shown in fig. 9, the fault locating device 90 includes a memory 901; and a processor 902 coupled to the memory 901, the memory 901 is used for storing instructions for executing the corresponding embodiment of the fault location method. The processor 902 is configured to perform the fault location method in any of the embodiments of the present disclosure based on instructions stored in the memory 901.
Fig. 10 illustrates a schematic diagram of a network management system according to some embodiments of the present disclosure.
As shown in fig. 10, the network management system 10 includes the fault locating device 100 in any of the above embodiments.
FIG. 11 illustrates a block diagram of a computer system for implementing some embodiments of the present disclosure.
As shown in FIG. 11, computer system 110 may take the form of a general purpose computing device. Computer system 110 includes a memory 1110, a processor 1120, and a bus 1100 that connects the various system components.
The memory 1110 may include, for example, system memory, non-volatile storage media, and the like. The system memory stores, for example, an operating system, an application program, a Boot Loader (Boot Loader), and other programs. The system memory may include volatile storage media, such as Random Access Memory (RAM) and/or cache memory. The non-volatile storage medium, for instance, stores instructions to perform corresponding embodiments of at least one of the fault localization methods. Non-volatile storage media include, but are not limited to, magnetic disk storage, optical storage, flash memory, and the like.
The processor 1120 may be implemented as discrete hardware components, such as a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gates or transistors, or the like. Accordingly, each of the modules such as the judging module and the determining module may be implemented by a Central Processing Unit (CPU) executing instructions in a memory to perform the corresponding steps, or may be implemented by a dedicated circuit to perform the corresponding steps.
Bus 1100 may employ any of a variety of bus architectures. For example, bus structures include, but are not limited to, Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, and Peripheral Component Interconnect (PCI) bus.
The computer system 110 may also include input-output interfaces 1130, network interfaces 1140, storage interfaces 1150, and the like. These interfaces 1130, 1140, 1150 and the memory 1110 and the processor 1120 may be connected by a bus 1100. The input/output interface 1130 may provide a connection interface for input/output devices such as a display, a mouse, and a keyboard. The network interface 1140 provides a connection interface for various networking devices. The storage interface 1150 provides a connection interface for external storage devices such as a floppy disk, a usb disk, and an SD card.
Various aspects of the present disclosure are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus and computer program products according to embodiments of the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions.
These computer-readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable apparatus to produce a machine, such that the execution of the instructions by the processor results in an apparatus that implements the functions specified in the flowchart and/or block diagram block or blocks.
These computer-readable program instructions may also be stored in a computer-readable memory that can direct a computer to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart and/or block diagram block or blocks.
The present disclosure may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects.
By the fault positioning method and device and the computer storage medium in the embodiment, the fault positioning efficiency is improved.
Thus far, the fault location method and apparatus, computer-readable storage medium according to the present disclosure have been described in detail. Some details that are well known in the art have not been described in order to avoid obscuring the concepts of the present disclosure. It will be fully apparent to those skilled in the art from the foregoing description how to practice the presently disclosed embodiments.

Claims (14)

1. A fault location method, comprising:
acquiring an alarm knowledge graph, wherein the alarm knowledge graph comprises entities and relations among the entities, and the entities comprise fault entities, root alarm entities generated by the fault entities and derivative alarm entities derived from the root alarm entities;
carrying out relation reasoning on the knowledge graph by using a graph neural network;
and determining the currently generated root alarm according to the result of the relationship inference and the current alarm data by using the classifier so as to position the fault.
2. The fault localization method of claim 1, wherein the using the graph neural network to perform relational inference on the knowledge graph comprises:
acquiring a characteristic vector of an entity in a knowledge graph as an initial vector of a node in a graph neural network;
and carrying out relationship reasoning according to the initial vector of the node in the graph neural network to obtain a final vector of the node, wherein the final vector of the node corresponds to the root alarm.
3. The fault location method according to claim 2, wherein the performing of the relationship inference according to the initial vectors of the nodes in the neural network of the graph to obtain the final vector of the nodes as a result of the relationship inference includes:
and determining the feature vector of the node at the next layer of the graph neural network according to the feature vector of the node at the current layer of the graph neural network and the feature vectors of other nodes having relations with the node at the current layer of the graph neural network until determining the feature vector of the node at the last layer of the graph neural network as the final vector of the node.
4. The fault location method of claim 1, wherein the determining, by the classifier, a currently occurring root alarm to locate the fault according to the result of the relational inference and the current alarm data comprises:
determining the current occurrence probability of one or more root alarms according to the result of the relationship inference and the current alarm data;
and taking the root alarm with the maximum probability as the current root alarm.
5. The fault location method of claim 1, wherein the determining, by the classifier, a currently occurring root alarm to locate the fault according to the result of the relational inference and the current alarm data comprises:
and positioning the fault according to the currently generated root alarm and the fault position information in the current alarm data.
6. The fault location method of claim 1, further comprising:
determining entities and the relation between the entities according to historical alarm data;
and establishing an alarm knowledge graph according to the entities and the relationship among the entities.
7. The method of claim 6, wherein the determining entities and relationships between entities based on historical alarm data comprises:
and determining the entities and the relationship among the entities according to the structured data in the historical alarm data.
8. The method of claim 6, wherein the determining the entities and the relationships between the entities based on the historical alarm data comprises:
and performing knowledge extraction on the semi-structured and unstructured historical alarm data to obtain entities and the relation between the entities.
9. The fault location method of claim 6, wherein the establishing an alarm knowledge graph according to the entities and relationships between the entities comprises:
carrying out knowledge fusion on the entities and the relation between the entities;
and establishing an alarm knowledge map according to the result of knowledge fusion.
10. The fault location method of claim 1, further comprising:
and reporting the positioned fault to a sub-slice management function, and processing the fault by the sub-slice management function.
11. A fault locating device comprises
The system comprises an acquisition module, a processing module and a display module, wherein the acquisition module is configured to acquire an alarm knowledge graph, the alarm knowledge graph comprises entities and relations among the entities, and the entities comprise fault entities, root alarm entities generated by the fault entities and derivative alarm entities derived from the root alarm entities;
an inference module configured to perform relational inference on the knowledge graph using a graph neural network;
and the positioning module is configured to determine a currently generated root alarm according to the result of the relation inference and the current alarm data by utilizing the classifier so as to position the fault.
12. A fault locating device comprising:
a memory; and
a processor coupled to the memory, the processor configured to perform the fault localization method of any of claims 1-10 based on instructions stored in the memory.
13. A network management system, comprising:
a fault location device as claimed in claim 11 or 12.
14. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method of fault localization as claimed in any one of claims 1 to 10.
CN202210453803.7A 2022-04-27 2022-04-27 Fault positioning method and device, and computer-storable medium Pending CN114785674A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210453803.7A CN114785674A (en) 2022-04-27 2022-04-27 Fault positioning method and device, and computer-storable medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210453803.7A CN114785674A (en) 2022-04-27 2022-04-27 Fault positioning method and device, and computer-storable medium

Publications (1)

Publication Number Publication Date
CN114785674A true CN114785674A (en) 2022-07-22

Family

ID=82433386

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210453803.7A Pending CN114785674A (en) 2022-04-27 2022-04-27 Fault positioning method and device, and computer-storable medium

Country Status (1)

Country Link
CN (1) CN114785674A (en)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109635117A (en) * 2018-12-26 2019-04-16 零犀(北京)科技有限公司 A kind of knowledge based spectrum recognition user intention method and device
CN111858123A (en) * 2020-07-29 2020-10-30 中国工商银行股份有限公司 Fault root cause analysis method and device based on directed graph network
CN112100369A (en) * 2020-07-29 2020-12-18 浙江大学 Semantic-combined network fault association rule generation method and network fault detection method
CN112269901A (en) * 2020-09-14 2021-01-26 合肥中科类脑智能技术有限公司 Fault distinguishing and reasoning method based on knowledge graph
US20210168021A1 (en) * 2019-11-30 2021-06-03 Huawei Technologies Co., Ltd. Fault Root Cause Determining Method and Apparatus, and Computer Storage Medium
CN113283027A (en) * 2021-05-20 2021-08-20 南京航空航天大学 Mechanical fault diagnosis method based on knowledge graph and graph neural network
US20220078071A1 (en) * 2020-04-07 2022-03-10 Huawei Technologies Co., Ltd. Device and method for monitoring communication networks
CN114221857A (en) * 2020-09-04 2022-03-22 中兴通讯股份有限公司 Fault locating method, electronic equipment and storage medium

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109635117A (en) * 2018-12-26 2019-04-16 零犀(北京)科技有限公司 A kind of knowledge based spectrum recognition user intention method and device
US20210168021A1 (en) * 2019-11-30 2021-06-03 Huawei Technologies Co., Ltd. Fault Root Cause Determining Method and Apparatus, and Computer Storage Medium
US20220078071A1 (en) * 2020-04-07 2022-03-10 Huawei Technologies Co., Ltd. Device and method for monitoring communication networks
CN111858123A (en) * 2020-07-29 2020-10-30 中国工商银行股份有限公司 Fault root cause analysis method and device based on directed graph network
CN112100369A (en) * 2020-07-29 2020-12-18 浙江大学 Semantic-combined network fault association rule generation method and network fault detection method
CN114221857A (en) * 2020-09-04 2022-03-22 中兴通讯股份有限公司 Fault locating method, electronic equipment and storage medium
CN112269901A (en) * 2020-09-14 2021-01-26 合肥中科类脑智能技术有限公司 Fault distinguishing and reasoning method based on knowledge graph
CN113283027A (en) * 2021-05-20 2021-08-20 南京航空航天大学 Mechanical fault diagnosis method based on knowledge graph and graph neural network

Similar Documents

Publication Publication Date Title
US11057266B2 (en) Identifying troubleshooting options for resolving network failures
US10693711B1 (en) Real-time event correlation in information networks
US10055275B2 (en) Apparatus and method of leveraging semi-supervised machine learning principals to perform root cause analysis and derivation for remediation of issues in a computer environment
CN110351150B (en) Fault source determination method and device, electronic equipment and readable storage medium
US9836952B2 (en) Alarm causality templates for network function virtualization
CN112580678A (en) Method and device for constructing cloud network alarm root relation tree model
Wang et al. Efficient alarm behavior analytics for telecom networks
CN113973042B (en) Method and system for root cause analysis of network problems
CN114202027B (en) Method for generating execution configuration information, method and device for model training
CN113657715A (en) Root cause positioning method and system based on kernel density estimation calling chain
CN113391943A (en) Micro-service fault root cause positioning method and device based on cause and effect inference
JP2023534696A (en) Anomaly detection in network topology
WO2020182051A1 (en) Method of recognizing a probable root cause alarm, apparatus and computer storage medium
CN112532408A (en) Method, device and storage medium for extracting fault propagation conditions
US9703924B2 (en) Timing constraints formulation for highly replicated design modules
CN114785674A (en) Fault positioning method and device, and computer-storable medium
TW202213096A (en) Device, method and computer program product for fault elimination of a multilayer system
CN115361295B (en) TOPSIS-based resource backup method, device, equipment and medium
CN115630073B (en) Electric power Internet of things data processing method and platform based on edge calculation
CN116938953A (en) Block chain-based data processing method and device, electronic equipment and storage medium
CN108768748B (en) Fault diagnosis method and device for power communication service and storage medium
CN116226256A (en) ETL flow visualization method, device, equipment and storage medium
US20230161661A1 (en) Utilizing topology-centric monitoring to model a system and correlate low level system anomalies and high level system impacts
CN112579402A (en) Method and device for positioning faults of application system
US11755609B2 (en) Automatically detecting data offloading methods using data bucketing and machine learning techniques

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination