A fault analysis method, apparatus, device, and medium

Abstract

This invention provides a fault analysis method, apparatus, device, and medium. The method includes: obtaining data of multiple feature indicators; determining the correlation between each pair of feature indicators based on their data; and, upon detecting that a first target feature indicator has triggered a first fault, determining a second target feature indicator that will trigger a second fault from among the multiple second target feature indicators, based on their respective data and those of multiple second target feature indicators that are positively correlated with the first target feature indicator. The first target feature indicator and the second feature indicator are each any one of the multiple feature indicators. This invention overcomes the limitations of rapid fault location in changing network environments by enabling iFIT to perform quantitative regression tracing of different service attributes and accurately predict subsequent faults.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of network technology and security technology, and in particular to a fault analysis method, device, equipment and medium. BACKGROUND

[0002] With the construction and development of new metropolitan area networks, the randomness and variability of network characteristics and bearing service paths put forward higher requirements for intelligent operation and maintenance. In the new metropolitan area network, the in-situ Flow Information Telemetry (iFIT) high-precision flow detection technology is widely used in the basic network. Through direct measurement of service packets, the real-time packet loss rate and nanosecond-level precision delay of network performance parameters are obtained. The detection technology has only 8 bytes of real service traffic overhead, and has the advantages of convenient deployment, high precision, etc. It is especially suitable for voice, video and other high-quality network service requirements for business operation state monitoring and fault rapid positioning.

[0003] At present, iFIT can be based on global voice, video and other high-quality network service requirements for business operation state monitoring and fault rapid positioning, but the change of network environment lacks a reliable algorithm to quantitatively regress and trace the source of different object attributes and different business attributes after iFIT makes fault rapid positioning, and to predict possible faults in the new metropolitan area network. SUMMARY

[0004] In view of the above problems, the embodiments of the present application provide a fault analysis method, device, equipment and medium in order to overcome the above problems or at least partially solve the above problems.

[0005] The first aspect of the embodiments of the present application discloses a fault analysis method, which comprises:

[0006] obtaining data of a plurality of characteristic indicators;

[0007] determining the correlation between each two characteristic indicators according to the data of each two characteristic indicators;

[0008] In the case where it is detected that a first target characteristic indicator causes a first fault, determining a second target characteristic indicator that will cause a second fault from a plurality of second target characteristic indicators according to the data of each of the plurality of second target characteristic indicators which are positively correlated with the first target characteristic indicator;

[0009] The first target characteristic indicator and the second characteristic indicator are any of the plurality of characteristic indicators.

[0010] Optionally, the method further comprises:

[0011] determining a normal variation rate range of each of the second target characteristic indicators according to historical normal data of each of the second target characteristic indicators in a case where no fault is triggered;

[0012] determining a second target characteristic indicator that is about to trigger a second fault from the second target characteristic indicators according to data of the second target characteristic indicators that are positively correlated with the first target characteristic indicator, including:

[0013] determining a variation rate of each of the second target characteristic indicators according to data of each of the second target characteristic indicators;

[0014] determining a second target characteristic indicator that is about to trigger a second fault from the second target characteristic indicators according to data of the second target characteristic indicators that are positively correlated with the first target characteristic indicator, including:

[0015] Optionally, the method further includes:

[0016] determining a variation rate of the first target characteristic indicator according to data of the first target characteristic indicator in a case where the first target characteristic indicator is detected to trigger a first fault;

[0017] determining a variation rate of each of the second target characteristic indicators according to data of each of the second target characteristic indicators;

[0018] determining a second target characteristic indicator that is about to trigger a second fault from the second target characteristic indicators according to data of the second target characteristic indicators that are positively correlated with the first target characteristic indicator, including:

[0019] determining a second target characteristic indicator that is about to trigger a second fault from the second target characteristic indicators according to data of the second target characteristic indicators that are positively correlated with the first target characteristic indicator, including:

[0020] Optionally, the obtaining of the data of the plurality of characteristic indicators includes:

[0021] obtaining a plurality of data of each of the plurality of characteristic indicators within a preset time period;

[0022] the determining of the correlation between each two characteristic indicators according to the data of each two characteristic indicators includes:

[0023] determining the correlation between each two characteristic indicators within a preset time period according to a plurality of data of each two characteristic indicators within the preset time period.

[0024] Optionally, after the data of the plurality of characteristic indicators is obtained, the method further includes standardizing the data of each characteristic indicator according to the following steps:

[0025] calculating a mean value and a standard deviation of the data of each characteristic indicator.

[0026] Based on the mean and standard deviation of the data of each feature index, standardized data of each feature index is obtained;

[0027] According to the data of each two feature indexes, the correlation between each two feature indexes is determined, including:

[0028] According to the standardized data of each two feature indexes, the correlation between each two feature indexes is determined.

[0029] Optionally, the plurality of feature indexes are divided into a first feature index set and a second feature index set, the feature indexes in the first feature index set are business performance feature indexes, and the feature indexes in the second feature index set are network element feature indexes.

[0030] Optionally, according to the data of each two feature indexes, the correlation between each two feature indexes is determined, including:

[0031] In a case where the fault analysis requirement is fault analysis within a feature index set, the correlation between each two feature indexes in the first feature index set is determined, or the correlation between each two feature indexes in the second feature index set is determined;

[0032] In a case where the fault analysis requirement is cross-feature index set fault analysis, the correlation between any two feature indexes is determined, the any two feature indexes are from the first feature set and / or the second feature set.

[0033] A second aspect of the embodiment of the present application discloses a fault analysis device, the device comprising:

[0034] An acquisition module is configured to obtain data of a plurality of feature indexes;

[0035] A correlation module is configured to determine the correlation between each two feature indexes according to the data of each two feature indexes;

[0036] A fault module is configured to, in a case where a first target feature index causes a first fault, determine a second target feature index that will cause a second fault from a plurality of second target feature indexes according to the data of the plurality of second target feature indexes that are positively correlated with the first target feature index; the first target feature index and the second feature index are any feature index in the plurality of feature indexes.

[0037] A third aspect of the embodiment of the present application discloses an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the fault analysis method as implemented by the first aspect of the embodiment when executed.

[0038] In a fourth aspect, the present application provides a computer readable storage medium having stored thereon computer programs / instructions which, when executed by a processor, implement the fault analysis method according to the first aspect.

[0039] The embodiments of the present application have the following advantages:

[0040] In the embodiments of the present application, in order to solve the problem that the non-quantitative evaluation standard for fault positioning and pre-judgment of the new metropolitan area network cannot trace back and regress according to the index data, a fault analysis method is provided. First, the data of a plurality of feature indexes in the network are obtained, then the correlation between each two feature indexes is determined according to the data of each two feature indexes, and finally, in the case that a first target feature index causes a first fault, a second target feature index that will cause a second fault is determined from a plurality of second target feature indexes according to the data of the plurality of second target feature indexes that are positively correlated with the first target feature index. Thus, the iFIT can make quantitative regression tracing of different service attributes after making rapid fault positioning in the changing network environment, and accurately predict the remaining faults that will occur. BRIEF DESCRIPTION OF DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the description of the embodiments of the present application. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0042] Figure 1 is a fault analysis method step flowchart provided by the embodiments of the present application;

[0043] Figure 2 is a method step flowchart for determining a second feature index causing a first fault provided by the embodiments of the present application;

[0044] Figure 3 is a structural schematic diagram of a fault analysis device provided by the embodiments of the present application. DETAILED DESCRIPTION

[0045] In order to make the above-mentioned purposes, features and advantages of the present application more apparent and easy to understand, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0046] The embodiment of the present application provides a fault analysis method, which comprises the following steps: Figure 1 Figure 1 A fault analysis method provided by the embodiment of the present application comprises the following steps S101-S103:

[0047] S101: Obtain data of a plurality of characteristic indexes.

[0048] In the embodiment, the characteristic indexes refer to characteristic indexes of iFIT service monitoring in the new metropolitan area network, and include characteristic indexes related to network characteristics and characteristic indexes related to service performance characteristics. The characteristic indexes related to network characteristics include CPU utilization, memory utilization, power supply state, voltage, current, number of carried services, number of network element level alarms, number of board card level alarms, number of protocol level alarms, number of port level alarms, port inflow traffic, port outflow traffic, port inflow bandwidth utilization, port outflow bandwidth utilization, etc. The characteristic indexes related to service performance characteristics include jitter, delay, packet loss rate, service inflow traffic, service outflow traffic, service outflow bandwidth utilization, service inflow bandwidth utilization, etc.

[0049] In an optional embodiment, the obtaining of the data of the plurality of characteristic indexes comprises: obtaining a plurality of data of each of the plurality of characteristic indexes in a preset time period, that is, obtaining the data of each characteristic index in the new metropolitan area network according to a preset time interval in a preset time period. For example, for a preset time period of 10 minutes, the data of each characteristic index in the new metropolitan area network is obtained according to a preset time interval (such as 10 seconds), and then 60 data of each characteristic index are obtained.

[0050] S102: Determine the correlation between each two characteristic indexes according to the data of each two characteristic indexes.

[0051] In the embodiment, the correlation between two characteristic indexes can be obtained by calculating a correlation coefficient between the two characteristic indexes. If the correlation coefficient is equal to 0, it indicates that the two characteristic indexes are not correlated. If the correlation coefficient is between 0 and 1, it indicates that the two characteristic indexes are positively correlated, that is, when one characteristic index data increases, the other characteristic index data also increases accordingly. If the correlation coefficient is between -1 and 0, it indicates that the two characteristic indexes are negatively correlated, that is, when one characteristic index data increases, the other characteristic index data decreases accordingly.

[0052] In an optional embodiment, the determining of the correlation between each two characteristic indexes according to the data of each two characteristic indexes comprises: determining the correlation between each two characteristic indexes in a preset time period according to a plurality of data of each of the two characteristic indexes in the preset time period.

[0053] ​In the embodiment, the correlation coefficient of the two characteristic indexes is calculated according to the data of each two characteristic indexes in a preset time period, and then the correlation of the two characteristic indexes is determined. Specifically, the correlation coefficient Corr(y1,2) is calculated by the correlation coefficient calculation formula, and the formula is as follows:

[0054]

[0055] wherein, y1 and y2 are characteristic indexes, y i1 represents the i th data of the characteristic index y1, y i2 represents the i th data of the characteristic index y2,

[0056] Step S103: In the case where the first target characteristic index causes the first fault, the second target characteristic index that will cause the second fault is determined from the plurality of second target characteristic indexes according to the real-time detection data of each of the plurality of second target characteristic indexes that are positively correlated with the first target characteristic index; the first target characteristic index and the second characteristic index are any of the plurality of characteristic indexes.

[0057] In the embodiment, when the first target characteristic index causes the first fault, there may be a second target characteristic index that causes the second fault due to the correlation between the characteristic indexes. Therefore, after detecting that the first target characteristic index causes the first fault, i.e., the iFIT detects the fault, the different service attributes in the new metropolitan area network are quantitatively regressed and traced, the real-time detection data of each of the plurality of second target characteristic indexes that are positively correlated with the first target characteristic index is obtained, and the second fault and the second target characteristic index that will occur are predicted according to the change of the data.

[0058] In an optional embodiment, determining the second target characteristic index that will cause the second fault includes steps A1 to A3:

[0059] Step A1: The normal change rate range of each of the plurality of second target characteristic indexes is determined according to the historical normal data of each of the plurality of second target characteristic indexes in the case where no fault is caused.

[0060] In the embodiment, after the first fault occurs, the historical normal data of each second target characteristic index is obtained from the memory, and the historical normal data is analyzed to determine the normal change rate range of the data of the second target characteristic index, wherein the change rate refers to the change amount of the data per unit time. The range of the change rate of each different second target characteristic index may be different.

[0061] Furthermore, in order to ensure the accuracy of the change rate range of each second characteristic indicator, historical normal data from multiple different times are obtained, and the normal change rate range of each second target characteristic indicator is determined based on the historical normal data from multiple different times.

[0062] Step A2: Determine the rate of change of each of the multiple second target feature indicators based on their respective data.

[0063] In this embodiment, after the first fault occurs, the second target characteristic index data is monitored to determine the range of the rate of change of the second characteristic index data. For example, for CPU utilization, if it is detected that the utilization rate changes from 20% to 50% within 5 seconds, then the rate of change of CPU utilization is 6% (i.e., the change per second is 6%).

[0064] Step A3: Identify the second target characteristic indicator whose rate of change exceeds its normal range as the second target characteristic indicator that is about to trigger the second fault.

[0065] In this embodiment, the real-time rate of change of each second target feature indicator determined in step A2 is compared with the normal rate of change range of each second target feature indicator determined in step A1. If the real-time rate of change of the second target feature indicator exceeds the normal rate of change range, then the second target feature indicator is determined to be the one that will trigger a second fault. For example, for CPU utilization, assuming its normal rate of change range is 0 to 5%, when the actual rate of change of CPU utilization exceeds 5%, then CPU utilization is determined to be the second target feature indicator that will trigger a second fault. In this embodiment, for iFIT service monitoring feature indicators, the correlation between each pair of feature indicators is determined based on the data of each feature indicator, thereby overcoming the challenges of rapid fault location in a changing network environment. After iFIT performs rapid fault location, it can perform quantitative regression tracing of different service attributes and accurately predict other faults that are about to occur.

[0066] In new metropolitan area networks, a fault is often caused by the combined effects of multiple network characteristics. For example, a user's traffic spike at a certain point in time may be due to disruptions from other services, excessively high or low equipment temperatures. When a first target characteristic is detected as causing a first fault, a second target characteristic is determined based on the correlation between the characteristic indicators.

[0067] In one alternative embodiment, such as Figure 2 As shown, based on the correlation of characteristic indicators in the new metropolitan area network, the second characteristic indicator that caused the first fault is determined, including steps B1 to B4:

[0068] Step B1: in the case of detecting that the first target feature index triggers the first fault, determining the change rate of the first target feature index according to the data of the first target feature index.

[0069] Step B2: determining the change rate of each of the plurality of second target feature indexes according to the data of each of the plurality of second target feature indexes.

[0070] Step B3: determining the second target feature index whose change rate matches the change rate of the first target feature index as the second target feature index causing the first fault.

[0071] Step B4: determining the second target feature index whose change rate does not match the change rate of the first target feature index as the second target feature index irrelevant to the occurrence of the first fault.

[0072] In this embodiment, the change rate of the first target feature index is determined according to the real-time data of the first target feature index, that is, the change amount of the first target feature index in a unit of time is determined. Then the change rate of each second target feature index related to the first target feature index is determined, that is, the change amount of the second target feature index in a unit of time is determined. Since the first target feature index and each second target feature index are related, the change rates of the first target feature index and each second target feature index are also related.

[0073] In this embodiment, the matching of the change rate of the second target feature index and the change rate of the first target feature index means that the change rate of the second target feature index and the change rate of the first target feature index satisfy the correlation relationship between them. For example, the change rate of the first target feature index A is n, the change rate of the second target feature index B is m, the first target feature index A and the second target feature index B present a positive correlation relationship (i.e. A=kB), then the change rate n and the change rate m satisfy the positive correlation relationship, when the change rate m of the second target feature index B changes, it satisfies the correlation relationship (i.e. matching) with the change rate n of the first target feature index A when the first fault occurs, then the second target feature index B is also the second target feature index triggering the first fault.

[0074] Because the units and meanings of the feature indicators are different, it is difficult to directly compare and analyze the correlation between different feature indicators, for example, the voltage of the device and the CPU utilization rate cannot be compared together. Therefore, after obtaining the data of the multiple feature indicators, the feature indicator data needs to be converted into standardized data to facilitate subsequent fault analysis based on the feature indicator data. Specifically, first, the obtained feature indicator data is converted into unified percentage data, for example, the voltage of the device is converted into a percentage between 0 and 100% according to whether it is working normally. Then, the feature indicator data converted into unified percentage is standardized to obtain standardized data.

[0075] Optionally, the data of each feature indicator is standardized, including:

[0076] calculating the mean and standard deviation of the data of each feature indicator;

[0077] obtaining the standardized data of each feature indicator based on the mean and standard deviation of the data of each feature indicator;

[0078] determining the correlation between each two feature indicators according to the data of each two feature indicators, including: determining the correlation between each two feature indicators according to the standardized data of each two feature indicators.

[0079] Specifically, the mean μ j and the standard deviation s j of the data of each feature indicator are respectively represented as:

[0080]

[0081] wherein y ij represents the i-th data of the feature indicator y j , and there are n data in total.

[0082] The standardized data of each feature indicator of each feature indicator is represented as:

[0083]

[0084] wherein, represents the standardized data corresponding to the data y ij , that is, the standardized data corresponding to the i-th data in the data of the feature indicator y j .

[0085] In the embodiment, all the characteristic index factor data are converted into standardized data, the correlation between each characteristic index is calculated based on the standardized data, so as to achieve the purpose of quantitatively evaluating the standard, simplifying the massive data, and facilitating the second fault prediction or analysis of the second target characteristic index causing the fault. In addition, by converting the characteristic index data into standardized data, the intelligent quantitative calculation of the new metropolitan area network fault can be realized, and the quantitative index data for intelligent fault positioning is provided.

[0086] In an optional embodiment, the plurality of characteristic indexes are divided into a first characteristic index set and a second characteristic index set, the characteristic indexes in the first characteristic index set are service performance characteristic indexes, and the characteristic indexes in the second characteristic index set are network element characteristic indexes.

[0087] Specifically, the characteristic indexes in the new metropolitan area network are divided into service performance characteristic indexes and network element characteristic indexes according to service performance and network characteristics, and then the first characteristic index set and the second characteristic index set are obtained.

[0088] In the embodiment, in order to improve the efficiency of fault analysis, different processing is performed according to the fault analysis requirements of the user. Since the fault analysis requirements are different, the correlation between each two characteristic indexes is determined according to the data of each two characteristic indexes, including the following two cases:

[0089] (1) In the case of characteristic index set fault analysis of the fault analysis requirement, the correlation between each two characteristic indexes in the first characteristic index set is determined, or the correlation between each two characteristic indexes in the second characteristic index set is determined.

[0090] In the embodiment, the fault analysis requirement of the characteristic index set fault analysis means that the fault analysis is only performed on the network side or only performed on the service side. Different processing is performed according to the fault analysis requirements of the user.

[0091] Specifically, if the fault analysis is performed on the service side, that is, the fault analysis is performed on the first characteristic index set, the fault analysis process includes: determining the correlation between each first characteristic index in the first characteristic set, in the case of detecting that the first target characteristic index causes the first fault, detecting the change rate of the data of each first characteristic index positively correlated with the first target characteristic index, and determining the target characteristic index that will cause the fault according to the change rate of the data of each first characteristic index.

[0092] If the network end is analyzed for failure, i.e., the second feature index set is analyzed for failure, the failure analysis process includes: determining the correlation between each second feature index, detecting the data change rate of each second feature index that is positively correlated with the second target feature index in the case of detecting that the second target feature index causes a second failure, and determining the target feature index that will cause the first failure according to the data change rate of each second feature index.

[0093] (2) In the case of cross-feature index set failure analysis, the correlation between any two feature indexes is determined, and the any two feature indexes are from the first feature set and / or the second feature set.

[0094] In this embodiment, the failure analysis requirement is cross-feature index set failure analysis, which means that the failure is analyzed in the network system, i.e., the network end and the service end together. Specifically, the cross-feature index set failure analysis includes: determining the correlation between any two feature indexes, detecting the data change rate of each second feature index that is positively correlated with the first target feature index in the case of detecting that the first target feature index causes a first failure, and determining the second target feature index that will cause the second failure from each second feature index according to the data change rate of each second feature index. Wherein, the first target feature index can be any feature index in the first feature index set or the second feature index set, and the second feature index can be any feature index in the first feature index set or the second feature index set.

[0095] In practical application, the failure analysis method in this embodiment is integrated into a new type of metropolitan network management failure intelligent diagnosis and prediction module, which fully supports subsequent new type of metropolitan network customer service subscription scenarios, and finally realizes the high network intelligence operation and maintenance of new type of metropolitan network from instruction execution, operation and maintenance, customer perception, diversified analysis of data collection, policy and rule decision and expert knowledge base update.

[0096] In this embodiment, a failure analysis method is proposed to address the problem of non-quantitative evaluation criteria for new type of metropolitan network failure positioning and prediction. First, the data of multiple feature indexes is obtained, then the correlation between each two feature indexes is determined according to the data of each two feature indexes, and finally in the case of detecting that the first target feature index causes a first failure, the second target feature index that will cause a second failure is determined from the multiple second target feature indexes according to the data of each second target feature index that is positively correlated with the first target feature index. Further, in the changing network environment, iFIT can make quantitative regression and traceability of different business attributes after rapid positioning of failure, and accurately predict the remaining failures that will occur.

[0097] Furthermore, this embodiment transforms the characteristic data of each indicator into standardized data, and performs fault analysis based on the correlation between the characteristics of each indicator, simplifying massive amounts of data and eliminating the need for operational experience to diagnose or detect faults in advance. Moreover, the method provided in this embodiment is based on existing network monitoring capabilities, monitoring basic network characteristic indicators such as packet loss and latency, and enabling fault analysis of network devices without requiring upgrades or modifications.

[0098] This invention also provides a fault analysis device, such as... Figure 3 As shown, Figure 3 This is a schematic diagram of a fault analysis device provided in an embodiment of the present invention. The device includes:

[0099] Module 31 is used to obtain data for multiple feature indicators;

[0100] Related module 32 is used to determine the correlation between each pair of feature indicators based on the data of each pair of feature indicators;

[0101] Fault module 33 is used to determine, in the event that a first fault is detected caused by a first target feature indicator, a second target feature indicator that is about to cause a second fault from the plurality of second target feature indicators based on the data of each of the plurality of second target feature indicators that are positively correlated with the first target feature indicator; the first target feature indicator and the second feature indicator are each any one of the plurality of feature indicators.

[0102] In an optional embodiment, the device further includes:

[0103] The first fault submodule is used to determine the normal change rate range of each of the multiple second target characteristic indicators based on their historical normal data under the condition that no fault is caused.

[0104] The second fault submodule is used to determine the rate of change of each of the multiple second target feature indicators based on their respective data.

[0105] The third fault submodule is used to identify second target characteristic indicators whose rate of change exceeds their normal rate of change range as second target characteristic indicators that are about to trigger a second fault.

[0106] In an optional embodiment, the device further includes:

[0107] The first change determination module is used to determine the rate of change of the first target feature indicator based on the data of the first target feature indicator when a first fault is detected caused by the first target feature indicator.

[0108] The second change determination module is used to determine the change rate of each of the plurality of second target feature indicators based on the data of each of the plurality of second target feature indicators;

[0109] The first change matching module is used to determine the second target feature indicator that matches the rate of change with the rate of change of the first target feature indicator as the second target feature indicator that causes the first fault to occur.

[0110] The second change matching module is used to identify second target feature indicators whose change rate does not match the change rate of the first target feature indicator, and to determine second target feature indicators that are irrelevant to the occurrence of the first fault.

[0111] In one alternative embodiment, the device module includes:

[0112] The first acquisition submodule is used to acquire multiple data points of each of the multiple feature indicators within a preset time period;

[0113] The first correlation submodule is used to determine the correlation between each pair of feature indicators within the preset time period based on multiple data points of each pair of feature indicators within the preset time period.

[0114] In one alternative embodiment, the device module includes:

[0115] The mean and standard deviation calculation module is used to calculate the mean and standard deviation of the data for each feature indicator;

[0116] The standard value calculation module is used to obtain standardized data for each feature indicator based on the mean and standard deviation of the data for each feature indicator;

[0117] The second correlation submodule is used to determine the correlation between each pair of feature indicators based on the standardized data of each pair of feature indicators.

[0118] In one optional embodiment, the relevant module includes:

[0119] The third related submodule is used to determine the correlation between every two feature indicators in the first feature indicator set, or to determine the correlation between every two feature indicators in the second feature indicator set, when the fault analysis requirement is fault analysis within the feature indicator set.

[0120] The fourth related submodule is used to determine the correlation between any two feature indicators when the fault analysis requirement is cross-feature indicator set fault analysis, wherein the two feature indicators come from the first feature set and / or the second feature set.

[0121] This invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the fault analysis method described in this invention.

[0122] This invention also provides a computer-readable storage medium storing a computer program / instructions thereon, which, when executed by a processor, implements the fault analysis method described in this invention.

[0123] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0124] Embodiments of the present invention are described with reference to flowchart illustrations and / or block diagrams of methods and apparatus according to embodiments of the present invention. It should 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 program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0125] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0126] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0127] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present invention.

[0128] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0129] The above provides a detailed description of the fault analysis method, apparatus, device, and medium provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method of failure analysis, characterized by, The method comprises: obtaining data of a plurality of characteristic indexes; determining the correlation between each two characteristic indexes according to the data of each two characteristic indexes; in the case that a first target characteristic index is detected to cause a first fault, determining a second target characteristic index that will cause a second fault from a plurality of second target characteristic indexes that are positively correlated with the first target characteristic index according to the data of each of the plurality of second target characteristic indexes; the first target characteristic index and the second target characteristic index are any of the plurality of characteristic indexes respectively; the method further comprises: determining the normal change rate range of each of the plurality of second target characteristic indexes according to the historical normal data of each of the plurality of second target characteristic indexes in the case that no fault is caused; determining the second target characteristic index that will cause a second fault from a plurality of second target characteristic indexes that are positively correlated with the first target characteristic index according to the data of each of the plurality of second target characteristic indexes, comprises: determining the change rate of each of the plurality of second target characteristic indexes according to the data of each of the plurality of second target characteristic indexes; determining the second target characteristic index that will cause a second fault from a plurality of second target characteristic indexes that are positively correlated with the first target characteristic index according to the data of each of the plurality of second target characteristic indexes, comprises: determining the change rate of each of the plurality of second target characteristic indexes according to the data of each of the plurality of second target characteristic indexes; determining the second target characteristic index that will cause a second fault from a plurality of second target characteristic indexes that are positively correlated with the first target characteristic index according to the data of each of the plurality of second target characteristic indexes, comprises: determining the second target characteristic index that will cause a second fault from a plurality of second target characteristic indexes that are positively correlated with the first target characteristic index according to the data of each of the plurality of second target characteristic indexes, comprises: determining the second target characteristic index that will cause a second fault from a plurality of second target characteristic indexes that are positively correlated with the first target characteristic index according to the data of each of the plurality of second target characteristic indexes, comprises:

2. The method of claim 1, wherein, the obtaining of the data of a plurality of characteristic indexes comprises: obtaining a plurality of data of each of the plurality of characteristic indexes within a preset time period; the determining of the correlation between each two characteristic indexes according to the data of each two characteristic indexes comprises: determining the correlation between each two characteristic indexes within a preset time period according to a plurality of data of each of the two characteristic indexes within the preset time period.

3. The method according to any of claims 1-2, characterized by, After obtaining the data of a plurality of characteristic indexes, the method further comprises standardizing the data of each characteristic index according to the following steps: calculating the mean value and the standard deviation of the data of each characteristic index; obtaining the standardized data of each characteristic index based on the mean value and the standard deviation of the data of each characteristic index; the determining of the correlation between each two characteristic indexes according to the data of each two characteristic indexes comprises: determining the correlation between each two characteristic indexes according to the standardized data of each two characteristic indexes.

4. The method according to any of claims 1 to 3, characterized in that, the plurality of characteristic indexes are divided into a first characteristic index set and a second characteristic index set, the characteristic indexes in the first characteristic index set are business performance characteristic indexes, and the characteristic indexes in the second characteristic index set are network element characteristic indexes.

5. The method of claim 4, wherein, According to the data of each two feature indicators, the correlation between each two feature indicators is determined, including: In the case that the fault analysis requirement is the fault analysis within the feature indicator set, the correlation between each two feature indicators in the first feature indicator set is determined, or the correlation between each two feature indicators in the second feature indicator set is determined; In the case that the fault analysis requirement is the fault analysis across the feature indicator sets, the correlation between any two feature indicators is determined, the any two feature indicators are from the first feature set and / or the second feature set.

6. A failure analysis apparatus characterized by comprising: The device comprises: an acquisition module, configured to obtain the data of a plurality of feature indicators; a correlation module, configured to determine the correlation between each two feature indicators according to the data of each two feature indicators; a fault module, configured to, in the case that a first target feature indicator is detected to cause a first fault, determine a second target feature indicator which is about to cause a second fault from a plurality of second target feature indicators according to the data of the plurality of second target feature indicators which are positively correlated with the first target feature indicator; the first target feature indicator and the second target feature indicator are any feature indicator in the plurality of feature indicators; the device further comprises: a first fault sub-module, configured to determine the normal change rate range of each of the plurality of second target feature indicators according to the historical normal data of each of the plurality of second target feature indicators in the case that no fault is caused; a second fault sub-module, configured to determine the change rate of each of the plurality of second target feature indicators according to the data of each of the plurality of second target feature indicators; a third fault sub-module, configured to determine the second target feature indicator whose change rate exceeds the normal change rate range of itself as the second target feature indicator which is about to cause the second fault; a first change determination module, configured to, in the case that a first target feature indicator is detected to cause a first fault, determine the change rate of the first target feature indicator according to the data of the first target feature indicator; a second change determination module, configured to determine the change rate of each of the plurality of second target feature indicators according to the data of each of the plurality of second target feature indicators; a first change matching module, configured to determine the second target feature indicator whose change rate matches the change rate of the first target feature indicator as the second target feature indicator which causes the first fault to occur; a second change matching module, configured to determine the second target feature indicator whose change rate does not match the change rate of the first target feature indicator as the second target feature indicator which is irrelevant to the occurrence of the first fault.

7. An electronic device, comprising: The computer program / instructions are executed by the processor to implement the fault analysis method according to any one of claims 1-5.

8. A computer readable storage medium having stored thereon computer programs / instructions, characterized in that, The computer program / instructions are executed by the processor to implement the fault analysis method according to any one of claims 1-5.

Images