CN110309009A - Situation-based operation and maintenance fault root cause positioning method, device, equipment and medium - Google Patents
Situation-based operation and maintenance fault root cause positioning method, device, equipment and medium Download PDFInfo
- Publication number
- CN110309009A CN110309009A CN201910421407.4A CN201910421407A CN110309009A CN 110309009 A CN110309009 A CN 110309009A CN 201910421407 A CN201910421407 A CN 201910421407A CN 110309009 A CN110309009 A CN 110309009A
- Authority
- CN
- China
- Prior art keywords
- alarm
- situation
- similarity
- warning information
- historical
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/079—Root cause analysis, i.e. error or fault diagnosis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/32—Monitoring with visual or acoustical indication of the functioning of the machine
- G06F11/324—Display of status information
- G06F11/327—Alarm or error message display
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Human Resources & Organizations (AREA)
- Software Systems (AREA)
- Evolutionary Biology (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Bioinformatics & Computational Biology (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Alarm Systems (AREA)
Abstract
The invention provides a situation-based operation and maintenance fault root cause positioning method, a situation-based operation and maintenance fault root cause positioning device, computer equipment and a storage medium, wherein the method comprises the following steps: acquiring alarm information corresponding to the alarm; inputting the alarm information into a machine learning model trained in advance to obtain a corresponding fault root; the process that the machine learning model determines the fault root cause according to the alarm information comprises the following steps: according to the alarm information, calculating the similarity between the alarm and a plurality of historical situations; determining a situation for searching a fault root cause of the alarm according to the similarity between the alarm and a plurality of historical situations; and calculating the importance of each warning source in the situation, and determining the fault root cause of the warning according to the importance of each warning source in the situation. The method does not need to have high requirements on operation and maintenance personnel, saves time and labor, considers the linkage effect of faults, performs global analysis on root cause positioning from the global angle and improves the accuracy of root cause positioning.
Description
Technical field
The present invention relates to fault location technology fields, and in particular to a kind of O&M failure root based on situation is because of positioning side
Method, device, computer equipment and storage medium.
Background technique
It will appear various failures during service operation, generally require experienced operation maintenance personnel and go to read largely
System alarm information, by domain knowledge, to failure root because carrying out analysis deduction, this solution is needed to operation maintenance personnel
There is higher requirement, it is larger to manpower and material resources consumption, simultaneously because the limitation of operation maintenance personnel professional domain knowledge frequently can lead to
Failure root because positioning result have certain deviation.
For this purpose, related technical personnel propose and a kind of search failure root because this in such a way that sequence analysis is abnormal
Method mainly carries out data exception analysis in the way of sequence analysis, that is, utilizes historical data training sequence prediction model, such as
Moving average model, LSTM sequential forecasting models etc., using model prediction value and the true data target of machine difference into
Row abnormality detection, when difference exceeds threshold value, i.e., it is believed that machine breaks down.This method is mainly for a specific machine
Index carries out abnormality detection, and when there is more machines, needs that different models is trained to go to detect simultaneously, all multi-models can consume greatly
The calculating power of amount；At the same time, all multi-models can also generate a large amount of wrong report, bring a large amount of manpower consumption to O&M.In addition,
Sequence analyzes the chain effect that abnormal technology does not consider failure, and the incidence relation modeling to more machinery compartments is simultaneously insufficient, cannot
Enough detect existing association between exception, thus can not from global angle to root because global analysis is done in positioning.
Summary of the invention
In order to solve the above-mentioned technical problem or it at least is partially solved above-mentioned technical problem, the present invention provides a kind of bases
In situation O&M failure root because of localization method, device, computer equipment and storage medium.
In a first aspect, the present invention provides a kind of O&M failure root based on situation because of localization method, comprising:
It obtains this and alerts corresponding warning information, the warning information is included in generated announcement during service operation
Alert source, alarm time and non-conformance description information；
The warning information is input in advance trained machine learning model with obtain corresponding failure root because；Its
In, the machine learning model according to the warning information determine the failure root because process include: according to the alarm believe
Breath calculates this alarm similarity between multiple historical contexts respectively；Wherein, it is included in correspondence in each historical context
Warning information corresponding to history alarm in historical time section；It is similar between multiple historical contexts respectively according to this alarm
Degree, determine find this alarm failure root because situation；Calculate the different degree of each alarm source in the situation, and according to
In the situation different degree of each alarm source determine this alarm failure root because.
Second aspect, the present invention provide a kind of O&M failure root based on situation because of positioning device, comprising:
Data obtaining module alerts corresponding warning information for obtaining this, and the warning information is included in business fortune
Alarm source caused by during row, alarm time and non-conformance description information；
Root is because of determining module, for being input to the warning information in machine learning model trained in advance with acquisition pair
The failure root answered because；Wherein, the machine learning model according to the warning information determine the failure root because process include:
According to the warning information, this alarm similarity between multiple historical contexts respectively is calculated；Wherein, each history feelings
It include the warning information corresponding to history alarm in corresponding historical time section in border；According to this alarm respectively with multiple history
Similarity between situation, determine find this alarm failure root because situation；Calculate each alarm source in the situation
Different degree, and according to the different degree of alarm source each in the situation determine this alarm failure root because.
The third aspect, the present invention provide a kind of computer equipment, including processor and storage on a memory and can located
The step of computer program run on reason device, the processor realizes the above method when executing the computer program.
Fourth aspect, the present invention provide a kind of computer readable storage medium, are stored thereon with computer program, the meter
The step of above method is realized when calculation machine program is executed by processor.
The present invention provides a kind of O&M failure root based on situation is because of localization method, device, computer equipment and storage
Medium obtains the warning information of this alarm first, then warning information is input in machine learning model, machine learning mould
Type according to warning information determine this alarm failure root because.In the whole process, it does not need operation maintenance personnel excessively to participate in, therefore
Do not need have very high requirement to operation maintenance personnel, and not only time saving and energy saving.Further, machine learning model is determining this
The failure root of alarm calculates this alarm similarity between multiple historical contexts respectively, then basis because during first
This alarm similarity between multiple historical contexts respectively, determine find the failure root of this alarm because situation, finally
In this situation determine failure root because.Cause this alarm failure root because be possible to be not this alert in alarm source,
The alarm source being likely to be in other alarms will lead to a series of different because the appearance of a failure has chain reaction
Often, it is thus possible to can cause multiple alarm, therefore the failure root for finding this alarm because when not merely in this alarm
Middle searching will also be found in other alarms, thus here according to similarity determine one for trouble-shooting root because situation,
Between the alarm in the situation be have it is certain associated, so finally determining failure root because accuracy can be relatively high.
As it can be seen that present application contemplates the chain effects of failure, that is, consider the association between multiple alarm sources, from global angle to root because
Global analysis is done in positioning, improves root because of the accuracy of positioning.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is flow diagram of the O&M failure root in the embodiment of the present application based on situation because of localization method；
Fig. 2 be in the embodiment of the present application machine learning model according to the warning information determine the failure root because process
Schematic diagram；
Fig. 3 is structural block diagram of the O&M failure root in the embodiment of the present application based on situation because of positioning device；
Fig. 4 is the structural block diagram of computer equipment in the embodiment of the present application.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
In a first aspect, the embodiment of the present application provides a kind of O&M failure root based on situation because of localization method, such as Fig. 1 institute
Show, this method comprises:
S100, this corresponding warning information of alarm is obtained, the warning information is produced during being included in service operation
Raw alarm source, alarm time and non-conformance description information；
It will be appreciated that warning information refers to the abnormal log information that business procedure generates in the process of running, usually wrap
Alarm source, alarm time, non-conformance description information are included, certainly can also include the information such as alarm level, monitoring programme type.
Wherein, alarm source, which refers to, there is abnormal machine, for example, the service in the background server cluster of certain shopping website
Device.There may be multiple alarm sources in the warning information once alerted, that is, it is abnormal to there is a situation where that more machines occur.
Wherein, non-conformance description information refers to the description information to abnormal conditions, for example, certain module (processor) in machine
Abnormal data information etc..
S200, the warning information is input in machine learning model trained in advance to obtain corresponding failure root
Cause；
It will be appreciated that this warning information alerted is input in machine learning model trained in advance, engineering
Practise model will export this alarm failure root because.
Wherein, the machine learning model according to the warning information determine the failure root because process include:
S210, according to the warning information, calculate this alarm similarity between multiple historical contexts respectively；Its
In, it include the warning information corresponding to history alarm in corresponding historical time section in each historical context；
It will be appreciated that the appearance of a usual failure has chain reaction, multiple tasks or machine are caused to occur abnormal
To issue a series of alarms, the situation of chain reaction is brought to be known as situation by failure, therefore situation is as caused by failure
A series of a kind of describing mode of alarms.That is, may include the warning information repeatedly alerted in a scene.
For example, analyze morning one day 9:00 occur alarm failure root because when, can choose the first seven of this day
All situations in it are as historical context, that is to say, that when analyzing this alarm, it is contemplated that all announcements in the first seven day
It is alert, that is, consider failure all in the first seven day.
In practical applications, step S210 can specifically comprise the following steps:
S211, according to the warning information, calculate this alarm with each historical context in each time history alarm it
Between co-occurrence map distance, local sensitivity Hash distance and alarm time it is poor；
Since there may be multiple alarms in a historical context, this alarm and the historical context are being calculated
When similarity, the similarity in this alarm and the historical context between alarm each time is calculated first；Then according to this announcement
Similarity in the alert and historical context between each alarm, calculates the similarity of this alarm and the historical context.For example, will
The average value of similarity in this alarm and the historical context between each alarm is as this alarm and the historical context
Similarity.
In the similarity in this alarm of calculating and a historical context between an alarm, it may be considered that a variety of fingers
It marks, for example, the co-occurrence map distance, alarm time in this alarm and the historical context between the alarm are poor, local sensitivity Hash
Distance (for example, simhash distance) etc..
Wherein, co-occurrence map distance is that the alarm source alerted according to this and alarm co-occurrence figure determine, and alert co-occurrence figure
Basic assumption is that the alarm often while occurred often has certain causality, therefore what is alerted in statistics a period of time is total to
Now relationship effectively can help situation to cluster.Alarm co-occurrence figure is made of the side of node and connecting node, and node can be with
It is either the former combination such as alarm source, alarm type, failure described in non-conformance description information, the weight on side is node
The inverse of co-occurrence number, the smaller association represented between alarm representated by two nodes of distance on side is got between two nodes
By force.The construction method of alarm co-occurrence figure is: one time window of setting is used to cache the alarm in this period, for example, by when
Between window be set as one month, then only consider the alarm of this month, alarm co-occurrence figure be alarm structure all in this month
It builds.It is then all alarm additions in new alarm and current window in co-occurrence alarm figure after receiving a new alarm
A line, and update the weight on all sides.So the co-occurrence map distance in this alarm and the historical context between the alarm is
This alarm added by while in the historical context alarm while the distance between, for connection two alert node between
The length on side.
Wherein, local sensitivity Hash distance is determined according to the non-conformance description information in warning information.Firstly, being retouched to abnormal
It states information and carries out entity extraction, obtain the entity in this alarm；Then this alarm is calculated using local sensitivity hash algorithm
In entity and historical context in local sensitivity Hash distance between the entity that alerts.Wherein, so-called entity refers to exception
The main information occurred in description information.Non-conformance description information be often Chinese and English mixing, and traditional participle tool often only
One kind that Chinese or English can be extracted, when Sino-British mixing, English is often verb and noun, and noun is often heavy
The entity wanted, therefore the combination that English word includes English, number and/or additional character is extracted, noun is retained as
Entity.Wherein, local sensitivity Hash distance is alerted in the entity and historical context in this alarm in non-conformance description information
Difference between entity is described, local sensitivity Hash distance is calculated by local sensitivity hash algorithm, and local sensitivity hash algorithm is logical
It is commonly used to find small text modification, and alarm caused by failure usually only has small difference, therefore local sensitivity Hash energy
It is enough effectively to find this kind of alarm.
As it can be seen that calculating the similarity between text entities by local sensitivity hash algorithm here, and alerts co-occurrence figure and fill
Divide the cooccurrence relation using alarm, compensates for the deficiency that local sensitivity hash algorithm only utilizes text information, facilitate from multiple
Angle carries out situation cluster.
S212, the co-occurrence map distance between each secondary history alarm in each historical context respectively is alerted according to this
It is poor from, local sensitivity Hash distance and alarm time, calculate the similarity between this alarm and the historical context.
Specific similarity calculation process can there are many, it is one of are as follows: for one threshold value of each single item setup measures and
Weighted value: co-occurrence map distance threshold value a1 and co-occurrence map distance weighted value a2, simhash distance threshold b1 and simhash distance power
Weight b2, alarm time difference threshold value c1 and alarm time weighted value c2；Then by this alarm with the historical context in the alarm it
Between co-occurrence map distance make the difference with co-occurrence map distance threshold value a1 and be multiplied again with co-occurrence map distance weighted value a2, by this alarm and should
Simhash distance in historical context between the alarm and simhash distance threshold b1 make the difference again with simhash distance weighting b2
Be multiplied, by this alarm and the historical context between the alarm alarm time and alarm time threshold value c1 make the difference again with alarm
Time weighting value c2 is multiplied, and finally three multiplied results are summed, the phase as this alarm and the alarm in the historical context
Like degree.A kind of this mode for only calculating similarity between alarm no longer arranges one by one here certainly there is also other modes
It lifts.
S220, the similarity between multiple historical contexts respectively is alerted according to this, determines the event for finding this alarm
Hinder root because situation；
It will be appreciated that so-called failure root causes the basic reason of failure because referring to, cause the failure root of this alarm
Because being possible to it is not the alarm source in this alarm, it may be possible to the alarm source in others alarm, because failure goes out
Chain reaction is now had, will lead to a series of exception, it is thus possible to multiple alarm can be caused, therefore finding this alarm
Failure root because when not merely found in this alarm, also to be found in other alarms.Here a feelings are determined first
Border, then in this situation trouble-shooting root because, be in fact it is some have associated alarm in trouble-shooting root because.
In practical applications, as shown in Fig. 2, step S220 may include steps of:
S221, the historical context for being greater than first threshold with the similarity of this alarm is judged whether there is: if it does not exist, then
A situation is created, this is alerted into corresponding warning information and is added in newly-built scene, and originally using newly-built situation as searching
The failure root of secondary alarm because situation.
It will be appreciated that illustrating this history feelings if the similarity between a historical context and this alarm is very big
Border and this alarm are closely similar, it is further contemplated that whether will in this historical context trouble-shooting root because.
Wherein, first threshold is determined in the training process of machine learning model, when according to side provided by the present application
Method finally determined by failure root because whether be true failure root because feedback, first threshold can be adjusted.
It will be appreciated that creating a situation, this is alerted into corresponding warning information and is added in newly-built scene, also
It is to say, only this alarm at present in newly-built situation.
It will be appreciated that there may be the historical context for being greater than first threshold with the similarity of this alarm, also having can
The historical context for being greater than first threshold with the similarity of this alarm can be not present.When it be not present, illustrate these historical contexts
Differed farther out with this alarm, thus only this alarm alarm source in trouble-shooting root because.
In practical applications, such when there is historical context of the similarity alerted with this greater than first threshold
Historical context is possible to only one, it is also possible to have it is multiple, when only one, can this alarm and this history
In situation trouble-shooting root because, that is to say, that as shown in Fig. 2, above-mentioned steps S220 can also include the following steps:
S222, the historical context for being greater than first threshold with the similarity of this alarm if it exists, then judgement and this alarm
Similarity be greater than whether the quantity of historical context of first threshold is 1: if so, warning information of this alarm be added to
The similarity of this alarm is greater than in the historical context of first threshold, and will add the historical context of the warning information of this alarm
As find this alarm failure root because situation.
It will be appreciated that when the similarity between only one historical context and this alarm is greater than first threshold, it is right
This historical context is updated, i.e., this warning information alerted is added in this historical context, and will be updated
Historical context as find below this alarm failure root because situation.Here to the update of historical context, actually and
It is a kind of alarm or situation cluster.
But it if there is the similarity between multiple historical contexts and this alarm is both greater than first threshold, then can choose
Wherein the higher historical context of similarity is merged out, in fused situation trouble-shooting root because.That is, as schemed
Shown in 2, above-mentioned steps S220 can also include the following steps:
If the quantity for the historical context that S223, the similarity alerted with this are greater than first threshold is greater than 1, this is accused
Alert warning information is added to be alerted in the highest historical context of similarity with this, and judges to add the alarm letter of this alarm
Whether the similarity between the historical context of breath and the historical context high with this alarm similarity time is greater than second threshold: if
It is then to melt the historical context for adding the warning information of this alarm and the historical context high with this alarm similarity time
Close, and using the obtained situation of fusion as find failure root that this is alerted because situation.
The highest historical context of similarity is alerted with this it will be appreciated that being added in the warning information for alerting this
In after, also to calculate the warning information of addition this alarm historical context and the history feelings time high with this alarm similarity
Similarity between border, and then judge whether the similarity between two situations is greater than second threshold.
Wherein, add this alarm warning information historical context add this alarm warning information before be with
This alarm highest historical context of similarity.Secondary height is only below highest, that is to say, that only accounts for two history feelings here
Border, one is that the highest historical context of similarity is alerted with this, and one is the historical context high with this alarm similarity time.
It will be appreciated that the main function of situation fusion is that similar situation is merged into a situation.Some situations are rigid
The alarm for starting to generate differs greatly, and can be clustered into multiple situations, over time, contextual content may tend to phase
Seemingly, this kind of situation is mainly merged in situation fusion, to reduce the workload of operation maintenance personnel debug.
If adding the historical context and the historical context high with this alarm similarity time of the warning information of this alarm
Between similarity it is very high, illustrate that the two situations are much like degree, therefore here merge the two situations, will merge
Situation later as find this alarm failure root because situation.But it if two situations are that much like degree is very low, says
Bright two situations are not suitable for fusion, and only trouble-shooting root is closed because comparing in the historical context of warning information for adding this alarm
It is suitable.That is, as shown in Fig. 2, above-mentioned steps S220 can also include the following steps:
If S224, the historical context for adding the warning information that this is alerted and the history feelings high with this alarm similarity time
Similarity between border is less than or equal to second threshold, then will add the historical context of the warning information of this alarm as searching
This alarm failure root because situation.
Wherein, the similarity between two situations is bigger, illustrates that the distance between two situations are closer, specifically can basis
The similarity between two situations of similarity calculation between alarm in two situations, and due to that may have multiple announcements in situation
It is alert, therefore can be using the minimum similarity between the alarm in a situation and the alarm between another situation as the two
Similarity between situation.
S230, the different degree for calculating each alarm source in the situation, and according to alarm source each in the situation
Different degree determine this alarm failure root because.
As soon as it will be appreciated that the alarm source in situation is more important, as failure root because a possibility that it is bigger.
In practical applications, Multiple factors can be considered when calculating the different degree of alarm source, for example, alarm source is in system
When PageRank value (rank value for being referred to as depended metrics of the alarm source in system calling figure) in calling figure, alarm
Between, alarm source generate frequency etc. of alarm.That is, above-mentioned steps S230 can be specifically included:
According to the frequency of the generation alarm of each alarm source, alarm time of origin in the situation and in system calling figure
In depended metrics rank value, calculate the different degree of the alarm source.
Wherein, system calling figure is true call graph in actual machine that business is arranged, in system calling figure
The alarm that the big node of importance generates is often also more important, therefore can more effectively indicate alarm using system calling figure
Importance.Few algorithms can make full use of system calling figure and carry out root because recommending in existing application.Alarm source is being
Ranking in calling figure of uniting is more forward, and the value is bigger, indicates that alarm source is more important, i.e. the higher node of PageRank value is corresponding
Alarm source be more likely to be failure root because.
Wherein, consider alarm time the reason of be: the more early alarm of generation time be more likely to be root because.Consider alarm source
Generate alarm frequency the reason of be: the corresponding alarm source of the lower node of alert frequency be more likely to be failure root because.
Due to when calculating the different degree of alarm source, it is contemplated that three kinds of factors, so this every kind factor all has a power
Weight values, weighted value can be determined in machine learning model training, be adjusted in machine learning model use process.Three
The sum of weighted value of factor principle is when there is new experience to be added, that is, to input a new warning information for 1
When, weighted value summation may change, at this time with regard to needing to be normalized, so that weighted value summation is still 1.
In practical application scene, unused business procedure, the granularity that different operation maintenance personnels divides failure is had
Very big difference, same set of parameter can not be suitable for all situations.At the same time, during program is run, this hair
It is bright parameter dynamically to be adjusted by the feedback of result, to reach better effect.Wherein, feedback as a result
Refer to, the alarm source that machine learning model is exported is real failure root because of the alarm that machine learning model is exported
The similarity or gap etc. of source and real failure root because between.Wherein, the parameter being related to have it is multiple, for example, first threshold,
Second threshold, the time window of co-occurrence figure, co-occurrence map distance threshold value, simHash threshold value, alarm time difference threshold value, root are because recommending
Normalized parameter or weighted value of Shi Butong experience etc..These parameters can obtain initial parameter by way of grid search.
In the process of running, can for root because the timely adjusting parameter of feedback, so that model be made to reach better effect.
O&M failure root provided by the present application obtains the warning information of this alarm because of localization method first, then will accuse
Alert information input into machine learning model, machine learning model according to warning information determine the failure root of this alarm because.?
It in whole process, does not need operation maintenance personnel and excessively participates in, therefore do not need have very high requirement to operation maintenance personnel, and not only save
Shi Shengli.Further, machine learning model determine this alarm failure root because during, calculate first this alarm divide
Then similarity not between multiple historical contexts alerts the similarity between multiple historical contexts respectively according to this,
Determine find this alarm failure root because situation, finally in this situation determine failure root because.Cause this alarm
Failure root is not the alarm source in this alarm because being possible to, it may be possible to the alarm source in others alarm, because of an event
The appearance of barrier has chain reaction, will lead to a series of exception, it is thus possible to can cause multiple alarm, therefore find this
The failure root of secondary alarm because when not merely found in this alarm, also to be found in other alarms, thus here according to
Similarity determine one for trouble-shooting root because situation, between the alarm in the situation be have it is certain associated, this
The failure root that sample finally determines because accuracy can be relatively high.As it can be seen that being considered present application contemplates the chain effect of failure
Association between multiple alarm sources improves root because of the accuracy of positioning from global angle to root because global analysis is done in positioning.Also
It is to say, more machines will be carried out while be considered by the application, and be no longer limited to single machine, mistake can be greatly decreased in this method
Root because recommending or early warning, reduce the cost of artificial investigation mistake, and the application is suitable for the failure under multiple application scenarios
Root is because of positioning.
Second aspect, the embodiment of the present application provide a kind of O&M failure root based on situation because of positioning device, such as Fig. 3 institute
Show, which includes:
Data obtaining module 310 alerts corresponding warning information for obtaining this, and the warning information is included in business
Generated alarm source, alarm time and non-conformance description information in operational process；
Root is because of determining module 320, for the warning information to be input in machine learning model trained in advance to obtain
Take corresponding failure root because；Wherein, the machine learning model according to the warning information determine the failure root because process
It include: that this alarm similarity between multiple historical contexts respectively is calculated according to the warning information；Wherein, each
It include the warning information corresponding to history alarm in corresponding historical time section in historical context；According to this alarm respectively with it is more
Similarity between a historical context, determine find this alarm failure root because situation；Calculate each in the situation
The different degree of alarm source, and according to the different degree of alarm source each in the situation determine this alarm failure root because.
That is, the machine learning model includes such as lower unit:
Similarity calculated, for according to the warning information, calculate this alarm respectively with multiple historical contexts it
Between similarity；It wherein, include the alarm letter corresponding to history alarm in corresponding historical time section in each historical context
Breath；
Situation determination unit is determined and is found for alerting the similarity between multiple historical contexts respectively according to this
This alarm failure root because situation；
Root is because of determination unit, for calculating the different degree of each alarm source in the situation, and according in the situation
The different degree of each alarm source determine this alarm failure root because.
In some embodiments, similarity calculated is specifically used for: according to the warning information, calculate this alarm with
The co-occurrence map distance between history alarm, local sensitivity Hash distance and alarm time are poor each time in each historical context；
The co-occurrence map distance between each secondary history alarm in each historical context, local sensitivity Hash respectively are alerted according to this
Distance and alarm time are poor, calculate the similarity between this alarm and the historical context.
In some embodiments, situation determination unit is specifically used for: judging whether there is big with the similarity of this alarm
In the historical context of first threshold: if it does not exist, then creating a situation, this is alerted corresponding warning information and is added to newly
Build in scene, and using newly-built situation as find this alarm failure root because situation.
In some embodiments, situation determination unit is specifically also used to: being greater than the with the similarity of this alarm if it exists
The historical context of one threshold value then judges to be greater than whether the quantity of the historical context of first threshold is 1 with the similarity of this alarm:
If so, the warning information of this alarm is added in the historical context for being greater than first threshold with the similarity of this alarm, and
Will add this alarm warning information historical context as find this alert failure root because situation.
In some embodiments, situation determination unit is specifically also used to: if the similarity with this alarm is greater than the first threshold
The quantity of the historical context of value is greater than 1, then this warning information alerted is added to similarity is highest goes through with this alarm
In history situation, and judges to add the historical context of the warning information of this alarm and alert the high history feelings of similarity time with this
Whether the similarity between border is greater than second threshold: if so, by add this alarm warning information historical context and with
The situation that this high historical context of alarm similarity time is merged, and fusion is obtained is as the failure for finding this alarm
Root because situation.
In some embodiments, situation determination unit is specifically also used to: if adding the history of the warning information of this alarm
Similarity between situation and the historical context high with this alarm similarity time is less than or equal to second threshold, then originally by addition
The historical context of the warning information of secondary alarm as find this alarm failure root because situation.
In some embodiments, root is specifically used for because of determination unit: according to the generation of each alarm source in the situation
Frequency, alarm time of origin and the depended metrics rank value in system calling figure of alarm, calculate the different degree of the alarm source.
The third aspect, the embodiment of the present application provide a kind of computer equipment, including memory, processor and are stored in storage
On device and the computer program that can run on a processor, which is characterized in that when the processor executes the computer program
The step of realizing the method that first aspect provides.
Fig. 4 shows the internal structure chart of computer equipment in one embodiment.As shown in figure 4, the computer equipment packet
Including the computer equipment includes processor, memory, network interface, input unit and the display screen connected by system bus
Deng.Wherein, memory includes non-volatile memory medium and built-in storage.The non-volatile memory medium of the computer equipment is deposited
Operating system is contained, computer program can also be stored with, when which is executed by processor, may make that processor is real
Now the O&M failure root based on situation is because of localization method.Computer program can also be stored in the built-in storage, the computer
When program is executed by processor, processor may make to execute the O&M failure root based on situation because of localization method.Computer equipment
Display screen can be liquid crystal display or electric ink display screen, the input unit of computer equipment can be on display screen
The touch layer of covering is also possible to the key being arranged on computer equipment shell, trace ball or Trackpad, can also be external
Keyboard, Trackpad or mouse etc..
It will be understood by those skilled in the art that structure shown in Fig. 4, only part relevant to application scheme is tied
The block diagram of structure does not constitute the restriction for the computer equipment being applied thereon to application scheme, specific computer equipment
It may include perhaps combining certain components or with different component layouts than more or fewer components as shown in the figure.
In one embodiment, the O&M failure root provided by the present application based on situation can be implemented as one because of positioning device
The form of kind computer program, computer program can be run in computer equipment as shown in Figure 4.The storage of computer equipment
The each program module for forming the positioning device can be stored in device, the computer program that each program module is constituted makes processor
The O&M failure root of each embodiment of the application described in this specification is executed because of the step in positioning.
Fourth aspect, the embodiment of the present application provide a kind of computer readable storage medium, are stored thereon with computer program,
The step of method that first aspect provides is realized when the computer program is executed by processor.
It will be appreciated that computer equipment and fourth aspect that the device of second aspect offer, the third aspect provide mention
The storage medium of confession is corresponding with the method that first aspect provides, the contents such as explanation, citing, beneficial effect in relation to content
Can be with reference to the corresponding portion in first aspect, details are not described herein again.
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to
Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.
The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although with reference to the foregoing embodiments
Invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each implementation
Technical solution documented by example is modified or equivalent replacement of some of the technical features；And these modification or
Replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution.
Claims (10)
1. a kind of O&M failure root based on situation is because of localization method characterized by comprising
It obtains this and alerts corresponding warning information, the warning information is included in generated alarm during service operation
Source, alarm time and non-conformance description information；
The warning information is input in advance trained machine learning model with obtain corresponding failure root because；
Wherein, the machine learning model according to the warning information determine the failure root because process include:
According to the warning information, this alarm similarity between multiple historical contexts respectively is calculated；Wherein, each is gone through
It include the warning information corresponding to history alarm in corresponding historical time section in history situation；
According to this alarm similarity between multiple historical contexts respectively, determine find the failure root of this alarm because feelings
Border；
The different degree of each alarm source in the situation is calculated, and is determined according to the different degree of alarm source each in the situation
This alarm failure root because.
2. calculating this alarm point the method according to claim 1, wherein described according to the warning information
Similarity not between multiple historical contexts, comprising:
According to the warning information, the co-occurrence figure between history alarm each time is calculated in this alarm and each historical context
Distance, local sensitivity Hash distance and alarm time are poor；
The co-occurrence map distance between each secondary history alarm in each historical context, local sensitivity respectively are alerted according to this
Hash distance and alarm time are poor, calculate the similarity between this alarm and the historical context.
3. the method according to claim 1, wherein it is described according to this alarm respectively with multiple historical contexts it
Between similarity, determine find this alarm failure root because situation, comprising:
Judge whether there is the historical context for being greater than first threshold with the similarity of this alarm:
If it does not exist, then a situation is created, this is alerted into corresponding warning information and is added in newly-built scene, and will be created
Situation as find this alarm failure root because situation.
4. according to the method described in claim 3, it is characterized in that, it is described according to this alarm respectively with multiple historical contexts it
Between similarity, determine find this alarm failure root because situation, further includes:
It is greater than the historical context of first threshold with the similarity of this alarm if it exists, then judges big with the similarity of this alarm
In the quantity of the historical context of first threshold whether be 1:
If so, this warning information alerted to be added to the historical context for being greater than first threshold with the similarity of this alarm
In, and will add this alarm warning information historical context as find this alert failure root because situation.
5. according to the method described in claim 4, it is characterized in that, it is described according to this alarm respectively with multiple historical contexts it
Between similarity, determine find this alarm failure root because situation, further includes:
If the quantity for being greater than the historical context of first threshold with the similarity of this alarm is greater than 1, the alarm that this is alerted
Information is added to be alerted in the highest historical context of similarity with this, and judges to add the history of the warning information of this alarm
Whether the similarity between situation and the historical context high with this alarm similarity time is greater than second threshold:
If so, by the historical context for the warning information for adding this alarm and the historical context high with this alarm similarity time
Merged, and using the obtained situation of fusion as find failure root that this is alerted because situation.
6. according to the method described in claim 5, it is characterized in that, it is described according to this alarm respectively with multiple historical contexts it
Between similarity, determine find this alarm failure root because situation, further includes:
If adding between the historical context of the warning information of this alarm and the historical context high with this alarm similarity time
Similarity is less than or equal to second threshold, then the historical context that will add the warning information of this alarm is alerted as this is found
Failure root because situation.
7. described in any item methods according to claim 1~6, which is characterized in that described to calculate each announcement in the situation
The different degree in alert source, comprising:
According to the frequency of the generation alarm of each alarm source, alarm time of origin in the situation and in system calling figure
Depended metrics rank value calculates the different degree of the alarm source.
8. a kind of O&M failure root based on situation is because of positioning device characterized by comprising
Data obtaining module alerts corresponding warning information for obtaining this, and the warning information is included in service operation mistake
Generated alarm source, alarm time and non-conformance description information in journey；
Root is corresponding to obtain in machine learning model trained in advance for the warning information to be input to because of determining module
Failure root because；Wherein, the machine learning model according to the warning information determine the failure root because process include: basis
The warning information calculates this alarm similarity between multiple historical contexts respectively；Wherein, in each historical context
Including warning information corresponding to the history alarm in corresponding historical time section；According to this alarm respectively with multiple historical contexts
Between similarity, determine find this alarm failure root because situation；Calculate the weight of each alarm source in the situation
Spend, and according to the different degree of alarm source each in the situation determine this alarm failure root because.
9. a kind of computer equipment including memory, processor and stores the meter that can be run on a memory and on a processor
Calculation machine program, which is characterized in that the processor realizes any one of claims 1 to 7 institute when executing the computer program
The step of stating method.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
The step of method described in any one of claims 1 to 7 is realized when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910421407.4A CN110309009B (en) | 2019-05-21 | 2019-05-21 | Situation-based operation and maintenance fault root cause positioning method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910421407.4A CN110309009B (en) | 2019-05-21 | 2019-05-21 | Situation-based operation and maintenance fault root cause positioning method, device, equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110309009A true CN110309009A (en) | 2019-10-08 |
| CN110309009B CN110309009B (en) | 2022-05-13 |
Family
ID=68075535
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910421407.4A Active CN110309009B (en) | 2019-05-21 | 2019-05-21 | Situation-based operation and maintenance fault root cause positioning method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110309009B (en) |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111158977A (en) * | 2019-12-12 | 2020-05-15 | 深圳前海微众银行股份有限公司 | Abnormal event root cause positioning method and device |
| CN111641519A (en) * | 2020-04-30 | 2020-09-08 | 平安科技（深圳）有限公司 | Abnormal root cause positioning method, device and storage medium |
| CN112003718A (en) * | 2020-09-25 | 2020-11-27 | 南京邮电大学 | Network alarm positioning method based on deep learning |
| CN112087334A (en) * | 2020-09-09 | 2020-12-15 | 中移（杭州）信息技术有限公司 | Alarm root cause analysis method, electronic device and storage medium |
| CN112181758A (en) * | 2020-08-19 | 2021-01-05 | 南京邮电大学 | Fault root cause positioning method based on network topology and real-time alarm |
| CN112866010A (en) * | 2021-01-04 | 2021-05-28 | 聚好看科技股份有限公司 | Fault positioning method and device |
| CN112953738A (en) * | 2019-11-26 | 2021-06-11 | ***通信集团山东有限公司 | Root cause alarm positioning system, method and device and computer equipment |
| WO2021179574A1 (en) * | 2020-03-12 | 2021-09-16 | 平安科技（深圳）有限公司 | Root cause localization method, device, computer apparatus, and storage medium |
| CN113407370A (en) * | 2020-03-16 | 2021-09-17 | ***通信有限公司研究院 | Root cause error clustering method, device, equipment and computer readable storage medium |
| CN113497716A (en) * | 2020-03-18 | 2021-10-12 | 华为技术有限公司 | Similar fault recommendation method and related equipment |
| CN113740666A (en) * | 2021-08-27 | 2021-12-03 | 西安交通大学 | Method for positioning storm source fault of data center power system alarm |
| CN113872780A (en) * | 2020-06-30 | 2021-12-31 | 大唐移动通信设备有限公司 | Fault root cause analysis method, device and storage medium |
| CN114090326A (en) * | 2022-01-14 | 2022-02-25 | 云智慧（北京）科技有限公司 | Alarm root cause determination method, device and equipment |
| WO2022057428A1 (en) * | 2020-09-18 | 2022-03-24 | 华为技术有限公司 | Method and apparatus for determining root cause of fault, and related device |
| CN114237962A (en) * | 2021-12-21 | 2022-03-25 | 中国电信股份有限公司 | Alarm root cause judgment method, model training method, device, equipment and medium |
| CN114325232A (en) * | 2021-12-28 | 2022-04-12 | 微梦创科网络科技（中国）有限公司 | Fault positioning method and device |
| CN114513802A (en) * | 2022-01-04 | 2022-05-17 | 武汉烽火技术服务有限公司 | Event stream-based bearer network fault analysis method and device |
| CN114742247A (en) * | 2022-04-08 | 2022-07-12 | 广东电网有限责任公司 | Characteristic extraction method and device based on distribution network distribution transformer abnormal alarm information |
| CN114944956A (en) * | 2022-05-27 | 2022-08-26 | 深信服科技股份有限公司 | Attack link detection method and device, electronic equipment and storage medium |
| CN115174251A (en) * | 2022-07-19 | 2022-10-11 | 深信服科技股份有限公司 | False alarm identification method and device for safety alarm and storage medium |
| CN116582410A (en) * | 2023-05-24 | 2023-08-11 | 青岛海信信息科技股份有限公司 | Intelligent operation and maintenance service method and device based on ITSM system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102638100A (en) * | 2012-04-05 | 2012-08-15 | 华北电力大学 | District power network equipment abnormal alarm signal association analysis and diagnosis method |
| WO2015051638A1 (en) * | 2013-10-08 | 2015-04-16 | 华为技术有限公司 | Fault location method and device |
| CN107770797A (en) * | 2016-08-17 | 2018-03-06 | ***通信集团内蒙古有限公司 | A kind of association analysis method and system of wireless network alarm management |
| CN108170702A (en) * | 2017-11-15 | 2018-06-15 | 国网河北省电力有限公司信息通信分公司 | A kind of power communication alarm association model based on statistical analysis |
-
2019
- 2019-05-21 CN CN201910421407.4A patent/CN110309009B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102638100A (en) * | 2012-04-05 | 2012-08-15 | 华北电力大学 | District power network equipment abnormal alarm signal association analysis and diagnosis method |
| WO2015051638A1 (en) * | 2013-10-08 | 2015-04-16 | 华为技术有限公司 | Fault location method and device |
| CN107770797A (en) * | 2016-08-17 | 2018-03-06 | ***通信集团内蒙古有限公司 | A kind of association analysis method and system of wireless network alarm management |
| CN108170702A (en) * | 2017-11-15 | 2018-06-15 | 国网河北省电力有限公司信息通信分公司 | A kind of power communication alarm association model based on statistical analysis |
Cited By (34)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112953738A (en) * | 2019-11-26 | 2021-06-11 | ***通信集团山东有限公司 | Root cause alarm positioning system, method and device and computer equipment |
| CN111158977B (en) * | 2019-12-12 | 2023-07-11 | 深圳前海微众银行股份有限公司 | Abnormal event root cause positioning method and device |
| CN111158977A (en) * | 2019-12-12 | 2020-05-15 | 深圳前海微众银行股份有限公司 | Abnormal event root cause positioning method and device |
| WO2021114977A1 (en) * | 2019-12-12 | 2021-06-17 | 深圳前海微众银行股份有限公司 | Method and device for positioning fundamental cause of abnormal event |
| WO2021179574A1 (en) * | 2020-03-12 | 2021-09-16 | 平安科技（深圳）有限公司 | Root cause localization method, device, computer apparatus, and storage medium |
| CN113407370A (en) * | 2020-03-16 | 2021-09-17 | ***通信有限公司研究院 | Root cause error clustering method, device, equipment and computer readable storage medium |
| CN113497716B (en) * | 2020-03-18 | 2023-03-10 | 华为技术有限公司 | Similar fault recommendation method and related equipment |
| US11757701B2 (en) | 2020-03-18 | 2023-09-12 | Huawei Technologies Co., Ltd. | Method for recommending similar incident, and related device |
| CN113497716A (en) * | 2020-03-18 | 2021-10-12 | 华为技术有限公司 | Similar fault recommendation method and related equipment |
| WO2021217855A1 (en) * | 2020-04-30 | 2021-11-04 | 平安科技（深圳）有限公司 | Abnormal root cause positioning method and apparatus, and electronic device and storage medium |
| CN111641519B (en) * | 2020-04-30 | 2022-10-11 | 平安科技（深圳）有限公司 | Abnormal root cause positioning method, device and storage medium |
| CN111641519A (en) * | 2020-04-30 | 2020-09-08 | 平安科技（深圳）有限公司 | Abnormal root cause positioning method, device and storage medium |
| CN113872780A (en) * | 2020-06-30 | 2021-12-31 | 大唐移动通信设备有限公司 | Fault root cause analysis method, device and storage medium |
| CN112181758A (en) * | 2020-08-19 | 2021-01-05 | 南京邮电大学 | Fault root cause positioning method based on network topology and real-time alarm |
| CN112181758B (en) * | 2020-08-19 | 2023-07-28 | 南京邮电大学 | Fault root cause positioning method based on network topology and real-time alarm |
| CN112087334A (en) * | 2020-09-09 | 2020-12-15 | 中移（杭州）信息技术有限公司 | Alarm root cause analysis method, electronic device and storage medium |
| WO2022057428A1 (en) * | 2020-09-18 | 2022-03-24 | 华为技术有限公司 | Method and apparatus for determining root cause of fault, and related device |
| CN114285730A (en) * | 2020-09-18 | 2022-04-05 | 华为技术有限公司 | Method and device for determining fault root cause and related equipment |
| CN112003718A (en) * | 2020-09-25 | 2020-11-27 | 南京邮电大学 | Network alarm positioning method based on deep learning |
| CN112866010A (en) * | 2021-01-04 | 2021-05-28 | 聚好看科技股份有限公司 | Fault positioning method and device |
| CN113740666A (en) * | 2021-08-27 | 2021-12-03 | 西安交通大学 | Method for positioning storm source fault of data center power system alarm |
| CN114237962A (en) * | 2021-12-21 | 2022-03-25 | 中国电信股份有限公司 | Alarm root cause judgment method, model training method, device, equipment and medium |
| CN114237962B (en) * | 2021-12-21 | 2024-05-14 | 中国电信股份有限公司 | Alarm root cause judging method, model training method, device, equipment and medium |
| CN114325232A (en) * | 2021-12-28 | 2022-04-12 | 微梦创科网络科技（中国）有限公司 | Fault positioning method and device |
| CN114325232B (en) * | 2021-12-28 | 2023-07-25 | 微梦创科网络科技（中国）有限公司 | Fault positioning method and device |
| CN114513802A (en) * | 2022-01-04 | 2022-05-17 | 武汉烽火技术服务有限公司 | Event stream-based bearer network fault analysis method and device |
| CN114513802B (en) * | 2022-01-04 | 2023-06-09 | 武汉烽火技术服务有限公司 | Method and device for analyzing bearing network faults based on event stream |
| CN114090326A (en) * | 2022-01-14 | 2022-02-25 | 云智慧（北京）科技有限公司 | Alarm root cause determination method, device and equipment |
| CN114742247A (en) * | 2022-04-08 | 2022-07-12 | 广东电网有限责任公司 | Characteristic extraction method and device based on distribution network distribution transformer abnormal alarm information |
| CN114944956A (en) * | 2022-05-27 | 2022-08-26 | 深信服科技股份有限公司 | Attack link detection method and device, electronic equipment and storage medium |
| CN115174251A (en) * | 2022-07-19 | 2022-10-11 | 深信服科技股份有限公司 | False alarm identification method and device for safety alarm and storage medium |
| CN115174251B (en) * | 2022-07-19 | 2023-09-05 | 深信服科技股份有限公司 | False alarm identification method and device for safety alarm and storage medium |
| CN116582410A (en) * | 2023-05-24 | 2023-08-11 | 青岛海信信息科技股份有限公司 | Intelligent operation and maintenance service method and device based on ITSM system |
| CN116582410B (en) * | 2023-05-24 | 2023-10-27 | 青岛海信信息科技股份有限公司 | Intelligent operation and maintenance service method and device based on ITSM system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110309009B (en) | 2022-05-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110309009A (en) | Situation-based operation and maintenance fault root cause positioning method, device, equipment and medium | |
| CN110928772B (en) | Test method and device | |
| US20190087294A1 (en) | Method for establishing fault diagnosis technique based on contingent Bayesian networks | |
| US8473263B2 (en) | Multi-infrastructure modeling and simulation system | |
| CN107066256B (en) | Object change model modeling method based on tense | |
| CN105095048A (en) | Processing method for alarm correlation of monitoring system based on business rules | |
| CN111539493B (en) | Alarm prediction method and device, electronic equipment and storage medium | |
| CN106164795B (en) | Optimization method for classified alarm | |
| CN112380089A (en) | Data center monitoring and early warning method and system | |
| CN112702184A (en) | Fault early warning method and device and computer-readable storage medium | |
| Wang et al. | A proactive approach based on online reliability prediction for adaptation of service-oriented systems | |
| CN114546765A (en) | Cluster monitoring method, system, device and medium | |
| Chen et al. | FRL-MFPG: Propagation-aware fault root cause location for microservice intelligent operation and maintenance | |
| CN111901156B (en) | Method and device for monitoring faults | |
| CN107679404A (en) | Method and apparatus for determining software systems potential risk | |
| Paulheim | Efficient semantic event processing: Lessons learned in user interface integration | |
| CN112445684A (en) | Real-time fault diagnosis and early warning method and device and computer storage medium | |
| Caroprese et al. | Handling preferences in P2P systems | |
| CN114358655A (en) | Method and device for generating recommendation scheme | |
| CN114584453A (en) | Fault analysis method and device of application system | |
| CN115525257A (en) | Micro-service construction method and device based on SVG technology | |
| Kpodjedo et al. | Not all classes are created equal: toward a recommendation system for focusing testing | |
| Caroprese et al. | Modeling cooperation in P2P data management systems | |
| CN107292027A (en) | A kind of bounded model checking method of the linear period invariant based on extension | |
| BR102016008054A2 (en) | system, method for execution by a system and computer readable storage device |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |