Specific embodiment
In order to make those skilled in the art better understand the technical solutions in the application, below in conjunction with this specification
Attached drawing in embodiment is clearly and completely described the technical solution in this specification embodiment, it is clear that described
Embodiment is merely a part but not all of the embodiments of the present application.Based on the embodiment in the application, this field
Those of ordinary skill's every other embodiment obtained without creative efforts, all should belong to the application
The range of protection.
The thought of this specification embodiment is, the alarm threshold of monitoring exceptional service is used in determining service monitoring system
When value, the prediction error of prediction model in service monitoring system is taken into account, identified alert threshold can be improved in this way
Accuracy, also, determine by way of automation the alert threshold that monitoring exceptional service is used in service monitoring system, also
The determination efficiency and accuracy of alert threshold can be improved.Based on this, this specification embodiment provides a kind of abnormal traffic
Monitoring method, device, equipment and storage medium, it is following to be described in detail one by one.
The method that this specification embodiment provides can be applied to the terminal devices such as computer, computer, i.e. this method is held
Row main body can be terminal device, specifically, executing subject can be the monitoring dress for the abnormal traffic being mounted on terminal device
It sets.
Fig. 1 is one of the method flow diagram of monitoring method of abnormal traffic that this specification embodiment provides, shown in FIG. 1
Method includes at least following steps:
Step 102, it obtains using prediction model to the index to be predicted of each sample business in the achievement data for setting the moment
The prediction data predicted, and, obtain index to be predicted truthful data corresponding to the setting moment of each sample business;
Wherein, above-mentioned prediction model is the model in service monitoring system for being predicted the index to be predicted of target service.
In general, in order to realize the monitoring to the abnormal traffic in business handled by operation system, in service monitoring system
In be provided with the prediction model predicted certain indexs to be predicted of business, service monitoring system can be set one or
Multiple prediction models, different prediction models predicts the achievement data of the different indexs of different business, therefore, to industry
Before business is monitored, need to carry out the training of prediction model using a large amount of business sample.
In this specification embodiment, can complete prediction model training after, to prediction model tested when
It waits and executes method provided by this specification embodiment.Therefore, mentioned in step 102 to sample business can be to prediction
The test sample that model is tested, it is of course also possible to be other sample business, this specification embodiment is limited not to this
It is fixed.
It in a specific embodiment, can be by the related data input prediction model of each sample business, to various kinds
Achievement data of this index to be predicted at the setting moment is predicted, and obtains output as a result, the output knot from prediction model
Fruit is then prediction data corresponding to each sample business.Wherein, the related data of above-mentioned each sample business can be the business
Index to be predicted corresponding true value at various moments, by true value input prediction model corresponding to each moment, thus
The index value at some setting moment after each moment is predicted.In addition, in this specification embodiment, due to using
Be sample business, therefore, the index to be predicted of each sample business truthful data corresponding to the setting moment be then it is known,
Therefore, index to be predicted true value corresponding to the setting moment of each sample business can be directly acquired.
For example, the related data of input prediction model can be portfolio, t2 moment handled by t1 moment operation system
Portfolio handled by operation system, portfolio handled by t3 moment operation system, need to predict t4 moment operation system institute
Therefore portfolio input prediction model corresponding to t1, t2 and t3 moment can be carried out t4 moment institute by the portfolio of processing
The prediction of corresponding portfolio.
Certainly, when being predicted using prediction model, the data inputted may be multiple, it is not limited to 3, this
Place is exemplary illustration, does not constitute the restriction to this specification embodiment.
Wherein, in this specification embodiment, above-mentioned prediction model can be shot and long term memory network (long short
Term memory, LSTM) prediction model.
In addition, it is necessary to explanation, above-mentioned index to be predicted can be the index arbitrarily predicted in business, example
It such as, can be handled portfolio, business funds total value etc..
Step 104, prediction data and truthful data according to corresponding to each sample business determines corresponding to each sample business
Prediction difference data.
In this specification embodiment, above-mentioned prediction data then refers to the prediction index to each sample business in setting
The predicted value that the achievement data at quarter is predicted, truthful data then refer to the index to be predicted of each sample business at the setting moment
Corresponding true value.
In above-mentioned steps 104, when determining prediction difference data corresponding to each sample business, then various kinds is calculated separately
The difference of predicted value corresponding to this business and true value, using the difference as prediction difference data corresponding to sample business.
For example, prediction data corresponding to sample business 1 is x1, truthful data corresponding to sample business 1 is y1, can be with
Prediction difference data corresponding to sample business 1 are calculated by following formula;
z1=| x1-y1|
Wherein, in above-mentioned formula, z1Indicate prediction difference data corresponding to sample business 1.
Step 106, based on prediction difference data corresponding to each sample business, determine that rule determines according to the threshold value of setting
Service monitoring system is used for the alert threshold being monitored to abnormal traffic.
In this specification embodiment, after having obtained prediction difference data corresponding to each sample business, then being based on should
Prediction difference data determine the alert threshold that monitoring exceptional service is used in service monitoring system.
In a specific embodiment, in above-mentioned steps 106, based on prediction difference number corresponding to each sample business
According to alert threshold corresponding to the determining service monitoring system of threshold value determination rule according to setting specifically comprises the following steps one
And step 2:
Step 1: determining the mean value and standard deviation of prediction difference data corresponding to each sample business;
Step 2: being based on above-mentioned mean value and standard deviation, alert threshold corresponding to service monitoring system is determined.
Specifically, in this specification embodiment, according to mean value and standard deviation, above-mentioned police can be calculated by following formula
Report threshold value;
T=μ+N* σ
Wherein, in above-mentioned formula, T indicates that alert threshold corresponding to service monitoring system, μ indicate mean value, and σ indicates mark
Quasi- poor, N indicates constant.
In this specification embodiment, when carrying out the adjustment of alert threshold, it is only necessary to carry out the adjustment of above-mentioned N value i.e.
Can, it is simple to operate.
In the specific implementation, the value of above-mentioned N can be 3, it is of course also possible to according to practical business demand to the value of N
It is adjusted, this specification embodiment is not defined the specific value of above-mentioned N.
The method that this specification embodiment provides for ease of understanding, it is following that the above-mentioned alert threshold provided will be provided
Specific method of determination.
For example, in the specific implementation, the number of used sample business is n, wherein n is positive integer, accessed
Prediction data corresponding to n sample business is denoted as x=[x1,x2,...,xn], wherein x1It is pre- corresponding to sample business 1
Measured data, x2For prediction data corresponding to sample business 2, xnIt is then prediction data corresponding to sample business n.N sample industry
The corresponding truthful data of business is denoted as y=[y1,y2,...,yn], wherein y1For truthful data corresponding to sample business 1, y2For
Truthful data corresponding to sample business 2, ynIt is then truthful data corresponding to sample business n, corresponding to n sample business
Prediction difference data are denoted as: z=[z1,z2,...,zn], wherein zi=| xi-yi|, the value of i is 1,2 ..., n, wherein z1For
Prediction difference data, z corresponding to sample business 12For prediction difference data, z corresponding to sample business 2nFor sample business n
Corresponding prediction difference data;
After prediction difference data corresponding to n sample business have been determined, then need to calculate n prediction difference data
Mean value and standard deviation, specifically, prediction difference data corresponding to n sample business can be calculated by following formula respectively
Mean value and standard deviation;
Wherein, in above-mentioned formula, μ indicates the mean value of prediction difference data corresponding to n sample business, and σ indicates n
The standard deviation of prediction difference data corresponding to sample business.
After calculating mean value and standard deviation by above-mentioned formula, then service monitoring system is calculated by formula T=μ+N* σ
In be used for monitoring exceptional service alert threshold.
Fig. 2 is the two of the method flow diagram of the monitoring method for the abnormal traffic that this specification embodiment provides, shown in Fig. 2
Method includes at least following steps:
Step 202, it obtains using prediction model to the index to be predicted of each sample business in the achievement data for setting the moment
The prediction data predicted, and, obtain index to be predicted truthful data corresponding to the setting moment of each sample business.
Step 204, prediction data and truthful data according to corresponding to each sample business determines corresponding to each sample business
Prediction difference data.
Step 206, the mean value and standard deviation of prediction difference data corresponding to each sample business are determined.
Step 210, it is based on above-mentioned mean value and standard deviation, determines what service monitoring system was used to be monitored abnormal traffic
Alert threshold.
The method that this specification embodiment provides, when determining the alert threshold for being monitored to abnormal traffic, meter
The prediction data of prediction model and the prediction difference data of truthful data are calculated, and according to the mean value and mean difference of prediction difference data
It determines alert threshold, in this way, the prediction error of prediction model is taken into account, improves the accurate of identified alert threshold
Property, and then improve the accuracy rate of service monitoring system monitoring exceptional service;In addition, also being realized in this specification embodiment
The automation that alert threshold arrives determines, in this way, accuracy and efficiency must compared with determining alert threshold by manual type
Raising is arrived.
Certainly, above describe the detailed process that the mean value and standard deviation according to prediction difference data determine alert threshold,
In addition to this, in this specification embodiment, alert threshold, example can also be determined according to the other parameters of prediction difference data
Such as, can the prediction difference data according to corresponding to each sample business median and median absolute deviation determine alarm threshold
Value, specifically comprises the following steps (1) and step (2);
Step (1), the median and median absolute deviation for determining prediction difference data corresponding to each sample business;
Step (2) is based on above-mentioned median and median absolute deviation, determines alarm threshold corresponding to service monitoring system
Value.
In the specific implementation, can by prediction difference data corresponding to each sample business according to data height sequence into
Row sequence, if the number of prediction difference data be odd number, then will be located in the middle after sequence a prediction difference data as
Median will then be located in the middle the flat of two prediction difference data if the number of prediction difference data is even number after sequence
Mean value is as median.
For ease of understanding, following to be illustrated citing.
For example, it is assumed that the quantity of sample business is 5, corresponding prediction difference data are respectively 100,128,97,
106,89, it can be according to sequence (it is of course also possible to according to sequence from low to high) from high to low to above-mentioned prediction difference number
According to being ranked up, the prediction difference data after sequence are as follows: 128,106,100,97,89, coming intermediate prediction difference data is
100, therefore, median is then 100.If the quantity of sample business is 6, corresponding prediction difference data are respectively 100,
128,97,106,89,117, it can be according to sequence (it is of course also possible to according to sequence from low to high) from high to low to above-mentioned
Prediction difference data are ranked up, the prediction difference data after sequence are as follows: and 128,117,106,100,97,89, due to pre- error of measurement
The number of Value Data is therefore even number comes there are two most intermediate prediction difference data, respectively 106 and 100,106 Hes
100 average is 103, and therefore, available median is 103.
Wherein, median absolute deviation (Median Absolute Deviation, MAD) is referred to as position in absolutely
Number is defined as the median that each data point (i.e. each prediction difference data) arrive the absolute deviation of median, as follows:
MAD=median (| zi-m|)
In above-mentioned formula, ziIndicate prediction difference data corresponding to i-th of sample business, m indicates each sample business
The median of corresponding prediction difference data, MAD indicate median absolute deviation.
For ease of understanding, following specific calculating process that will illustrate above-mentioned median absolute deviation.
For example, continue to use the example above, the quantity of sample business is 5, corresponding prediction difference data are respectively 100,
128,97,106,89, median 100, the absolute deviation of each prediction difference data to median is respectively 0,28,3,6,11,
Median corresponding to each absolute deviation is then 6, and therefore, corresponding median absolute deviation is then 6.
Specifically, being based on median and median absolute deviation in above-mentioned steps (2), determining above-mentioned business monitoring system
The corresponding alert threshold of system, specifically includes:
According to median and median absolute deviation, above-mentioned alert threshold is calculated by following formula;
T=m+N*MAD
Wherein, in above-mentioned formula, T indicates that alert threshold corresponding to service monitoring system, m indicate median, MAD table
Show median absolute deviation, N indicates constant.
In the specific implementation, the value of above-mentioned N can be 3, it is of course also possible to according to practical business demand to the value of N
It is adjusted, this specification embodiment is not defined the specific value of above-mentioned N.
Fig. 3 is the three of the method flow diagram of the monitoring method for the abnormal traffic that this specification embodiment provides, shown in Fig. 3
Method includes at least following steps:
Step 302, it obtains using prediction model to the index to be predicted of each sample business in the achievement data for setting the moment
The prediction data predicted, and, obtain index to be predicted truthful data corresponding to the setting moment of each sample business.
Step 304, prediction data and truthful data according to corresponding to each sample business determines corresponding to each sample business
Prediction difference data.
Step 306, the median and median absolute deviation of prediction difference data corresponding to each sample business are determined.
Step 308, it is based on above-mentioned median and median absolute deviation, determines alarm threshold corresponding to service monitoring system
Value.
In addition, in this specification embodiment, in addition to can the prediction difference data according to corresponding to each sample business
Mean value and standard deviation, and the median and median absolute deviation of the prediction difference data according to corresponding to each sample business are true
Surely except for the alert threshold of monitoring exceptional service, it is also based on setting for prediction difference data corresponding to each sample business
Determine percentile and determines above-mentioned alert threshold.
Therefore, in above-mentioned steps 106, based on prediction difference data corresponding to each sample business, according to the threshold value of setting
Determine the alert threshold that rule determines that service monitoring system is used to be monitored abnormal traffic, further includes:
Calculate the setting percentile of prediction difference data corresponding to each sample business;The setting percentile is determined
For alert threshold corresponding to service monitoring system.
In this specification embodiment, due to having multiple sample business, available corresponding multiple prediction differences
Data are properly termed as prediction difference data sequence.In the specific implementation, can by the prediction difference data sequence according to from it is small to
Big sequence is ranked up, and then calculates the setting percentile of the prediction difference data sequence, will be corresponding to obtained serial number
Prediction difference data be determined as alert threshold.For example, in the specific implementation, calculating the 95%00 of the prediction difference data sequence
Any percentiles such as quantile, 80% percentile, can specifically be configured according to actual needs, and this specification is implemented
Example is defined not to this.
Certainly, in some cases, the setting percentile of prediction difference data sequence calculated is not integer, such as
The calculated percentile that set of institute is 2.56, at this moment will then be adjoined and greater than the determination of prediction difference data corresponding to the serial number
For alert threshold, i.e., third prediction difference data are determined as alert threshold.
For ease of understanding, following to be illustrated citing.
For example, in a specific embodiment, obtained prediction difference data sequence are as follows: 156,176,145,124,
132,109,112,100,107,109,89,98,132,115,131,145,123,116,106,107;According to from small to large
Sequence above-mentioned prediction difference data are ranked up, the prediction difference data sequence after obtained sequence are as follows: 89,98,100,
106,107,107,109,109,112,115,116,123,124,131,132,132,145,145,156,176;If pre- with this
95% percentile of error of measurement Value Data is as alert threshold, then serial number corresponding to 95 percentiles of the prediction difference data
For 20*95%=19, i.e., the 19th prediction difference data after above-mentioned sequence are determined as alert threshold, i.e., the alert threshold is
156;If further for example, prediction difference sequence are as follows: 156,176,145,124,132,109,112,100,107,109,89,98,
132,115,131,145;After being ranked up according to sequence from small to large to above-mentioned prediction difference data, obtained pre- error of measurement
Value Data are as follows: 89,98,100,107,109,109,112,115,124,131,132,132,145,145,156,176;If with this
95% percentile of prediction difference data is as alert threshold, then sequence corresponding to 95 percentiles of the prediction difference data
Number be 15*95%=14.25, then the 15th prediction difference data after above-mentioned sequence are determined as alert threshold, i.e. the alarm
Threshold value is 176.
Certainly, how above-mentioned be merely illustrative calculates the setting percentiles of prediction difference data and in addition to this passes through
Other modes, which calculate percentile, can be applied to this specification embodiment, no longer repeat various calculating settings hundred one by one herein
The specific implementation of quantile.
Certainly, the prediction difference data according to corresponding to each sample business are described respectively in this specification embodiment
Mean value and standard deviation, the median of the prediction difference data according to corresponding to each sample business and median absolute deviation, and,
The alarm threshold for being used for monitoring exceptional service is determined according to the setting percentile of prediction difference data corresponding to each sample business
The detailed process of value still determines that the detailed process of alert threshold is not limited thereto, in addition to this it is possible to according to various kinds
The other parameters of prediction difference data corresponding to this business determine alert threshold, and this specification embodiment will not enumerate.
The method that this specification embodiment provides for ease of understanding, it is following to handle certain using index to be predicted as operation system
The portfolio of business for calculating alert threshold by mean value and standard deviation, introduces the method that this specification embodiment provides.
Fig. 4 is the four of the method flow diagram of the monitoring method for the abnormal traffic that this specification embodiment provides, shown in Fig. 4
Method includes at least following steps:
Step 402, the prediction industry that the portfolio to each sample business at the setting moment is predicted is obtained from prediction model
Business amount, and, each sample business is obtained in the actual services amount at setting moment.
Step 404, the difference for calculating separately prediction portfolio and actual services amount corresponding to each sample business obtains each
Portfolio difference corresponding to sample business.
Step 406, mean value and standard deviation corresponding to each portfolio difference are calculated.
Step 408, calculate setting multiple standard deviation and mean value and value, this and value are determined as alert threshold so that
The service monitoring system is monitored abnormal traffic according to the alert threshold.
The monitoring method for the abnormal traffic that this specification embodiment provides, for abnormal industry in determining service monitoring system
When the alert threshold of business monitoring, the achievement data based on prediction model to the index to be predicted of sample business at the setting moment is carried out
Prediction of the index to be predicted of the prediction data of prediction and the sample business between the truthful data corresponding to the setting moment
Difference data is determined;In this way, the prediction error of the prediction model in service monitoring system is taken into account, improves and determine
Alert threshold accuracy, so as to improve the accuracy rate of monitoring exceptional service;In addition, this specification embodiment realizes
The automation of alert threshold determines that compared with determining alert threshold by manual type, efficiency and accuracy are all improved.
Corresponding to the monitoring method for the abnormal traffic that embodiment corresponding to this specification Fig. 1-Fig. 4 provides, based on identical
Thinking, this specification embodiment additionally provide a kind of monitoring method of abnormal traffic, and this method is applied to service monitoring system, i.e.,
The executing subject of this method is service monitoring system, is specifically the monitoring for the abnormal traffic being set on service monitoring system
Device.Fig. 5 is the five of the method flow diagram of the monitoring method for the abnormal traffic that this specification embodiment provides, side shown in fig. 5
Method includes at least following steps:
Step 502, the actual services data of operation system to be monitored target service handled by the setting moment are obtained, with
And it obtains prediction model and treats the prediction business number that monitoring business system target service handled by the setting moment is predicted
According to.
Wherein, above-mentioned actual services data can be operation system to be monitored target service handled by the setting moment
The data such as portfolio or the business amount of money, specifically, can determine according to actual needs, this specification is not to above-mentioned
The specific data of actual services data are defined.
It in the specific implementation, can be from business system to be monitored during operation system processing target business to be monitored
System obtains its generated actual services data, it can also be obtained from database corresponding to operation system to be monitored and is being set
Generated actual services data are carved in timing.Furthermore it is possible to obtain history from database corresponding to operation system to be monitored
Actual services data corresponding to each moment, and by actual services data incoming traffic monitoring system corresponding to each moment
Prediction model in, with by prediction model to above-mentioned operation system to be monitored setting the moment handled by target service industry
Business data are predicted, to obtain prediction business datum.
In addition, it is necessary to which explanation, above-mentioned acquired actual services data and prediction business datum are target service
Some operational indicator corresponding to business datum.For example, acquired actual services data and prediction business datum are mesh
Mark business is in the business total value or target service for setting the moment in the portfolio etc. for setting the moment.
Specifically, above-mentioned prediction model can be LSTM prediction model, it is, of course, also possible to be other prediction models, this theory
Bright book embodiment is defined not to this.
Step 504, the prediction difference data of above-mentioned prediction business datum and above-mentioned actual services data are determined.
Step 506, above-mentioned prediction difference data are compared with predetermined alert threshold, to determine target service
Whether there is exception at the setting moment;Wherein, alert threshold based on prediction model to each sample business given time business
Prediction difference of the prediction data and each sample business that data are predicted between the actual services data of above-mentioned given time
Data are determined.
In the specific implementation, it can be when above-mentioned prediction difference data are greater than above-mentioned alert threshold, then it is assumed that the moment
The processing of target service occurs abnormal, at this moment, can sound an alarm, for example, can send mail, short to relevant staff
Letter, prompt information etc., to handle in time.
It should be noted that above-mentioned alert threshold can be the test rank in prediction model in this specification embodiment
Determined by section;Above-mentioned each sample business for determining alert threshold can be each test specimens tested prediction model
This.
In the specific implementation, available prediction model to each test sample setting the moment operational indicator to be predicted into
The prediction data of row prediction, and truthful data of the operational indicator to be predicted of each test sample at the setting moment is obtained, it calculates
The prediction difference data of each prediction data and truthful data, then, based on pre- error of measurement corresponding to obtained each forecast sample
Value Data determines above-mentioned alert threshold.
Wherein, in this specification embodiment, above-mentioned alert threshold is based on pre- error of measurement corresponding to each sample business
Determined by the mean value and standard deviation of Value Data, it can specifically determine as follows:
Determine the mean value and standard deviation of prediction difference data corresponding to each sample business;Based on above-mentioned mean value and standard
Difference determines above-mentioned alert threshold.
Specifically, above-mentioned mean value and standard deviation can be based in this specification embodiment, determined by following formula
State alert threshold:
T=μ+N* σ
Wherein, in above-mentioned formula, T indicates that alert threshold, μ indicate mean value, and σ indicates standard deviation, and N indicates constant.
In another embodiment, alert threshold is based on prediction difference data corresponding to each sample business
Determined by median and median absolute deviation, it can specifically determine as follows:
Determine the median and median absolute deviation of prediction difference data corresponding to each sample business;Based among the above
Digit and median absolute deviation, determine above-mentioned alert threshold.
Specifically, in this specification embodiment above-mentioned median and median absolute deviation can be based on, by as follows
Formula determines above-mentioned alert threshold:
T=m+N*MAD
Wherein, in above-mentioned formula, T indicates that alert threshold, m indicate that median, MAD indicate median absolute deviation, N table
Show constant.
Certainly, in this specification embodiment, above-mentioned alert threshold is also based on prediction corresponding to each sample business
The setting percentile of difference data is determined.
It should be noted that the specific implementation process of above-mentioned each step can refer to embodiment of the method corresponding to Fig. 1-Fig. 4
In each step specific implementation process, details are not described herein again.
The monitoring method for the abnormal traffic that this specification embodiment provides is used when being monitored to abnormal traffic
Alert threshold, then be that the achievement data based on prediction model to the operational indicator to be predicted of sample business at the setting moment carries out
The operational indicator to be predicted of the prediction data of prediction and the sample business is between the truthful data corresponding to the setting moment
Prediction difference data are determined;In this way, the prediction error of the prediction model in service monitoring system is taken into account, institute is improved
The accuracy of determining alert threshold, so as to improve the accuracy rate of monitoring exceptional service.
Corresponding to the monitoring method for the abnormal traffic that embodiment corresponding to this specification Fig. 1-Fig. 4 provides, based on identical
Thinking, this specification embodiment additionally provide a kind of monitoring device of abnormal traffic, right for executing this specification Fig. 1-Fig. 4 institute
The method for answering embodiment, Fig. 6 are the module composition schematic diagram of the monitoring device for the abnormal traffic that this specification embodiment provides, figure
Device shown in 6 includes:
Module 602 is obtained, for obtaining the index to be predicted for using prediction model to each sample business at the setting moment
The prediction data that achievement data is predicted, and, the index to be predicted for obtaining each sample business was set corresponding to the moment
Truthful data;Wherein, prediction model is the mould in service monitoring system for being predicted the index to be predicted of target service
Type;
First determining module 604 determines various kinds for prediction data and truthful data according to corresponding to each sample business
Prediction difference data corresponding to this business;
Second determining module 606, for based on prediction difference data corresponding to each sample business, according to the threshold value of setting
Determine the alert threshold that rule determines that service monitoring system is used to be monitored abnormal traffic.
Optionally, above-mentioned second determining module 606, comprising:
First determination unit, for determining the mean value and standard deviation of prediction difference data corresponding to each sample business;
Second determination unit determines alert threshold corresponding to service monitoring system for being based on mean value and standard deviation.
Optionally, above-mentioned second determination unit, is specifically used for:
According to mean value and standard deviation, alert threshold is calculated by following formula;
T=μ+N* σ
Wherein, in above-mentioned formula, T indicates that alert threshold corresponding to service monitoring system, μ indicate mean value, and σ indicates mark
Quasi- poor, N indicates constant.
Optionally, above-mentioned second determining module 606, further includes:
Third determination unit, the median and median for determining prediction difference data corresponding to each sample business are exhausted
To deviation;
4th determination unit determines corresponding to service monitoring system for being based on median and median absolute deviation
Alert threshold.
Optionally, above-mentioned 4th determination unit, is specifically used for:
According to median and median absolute deviation, alert threshold is calculated by following formula;
T=m+N*MAD
Wherein, in above-mentioned formula, T indicates that alert threshold corresponding to service monitoring system, m indicate median, MAD table
Show median absolute deviation, N indicates constant.
Optionally, above-mentioned second determining module 606, further includes:
Computing unit, for calculating the setting percentile of prediction difference data corresponding to each sample business;
5th determination unit, for above-mentioned setting percentile of stating to be determined as alarm threshold corresponding to service monitoring system
Value.
Optionally, above-mentioned prediction model is LSTM prediction model.
The monitoring device of the abnormal traffic of this specification embodiment can also carry out the monitoring device of abnormal traffic in Fig. 1-Fig. 4
The method of execution, and the monitoring device of abnormal traffic is realized in Fig. 1-embodiment illustrated in fig. 4 function, details are not described herein.
The monitoring device for the abnormal traffic that this specification embodiment provides, for abnormal industry in determining service monitoring system
When the alert threshold of business monitoring, the achievement data based on prediction model to the index to be predicted of sample business at the setting moment is carried out
Prediction of the index to be predicted of the prediction data of prediction and the sample business between the truthful data corresponding to the setting moment
Difference data is determined;In this way, the prediction error of the prediction model in service monitoring system is taken into account, improves and determine
Alert threshold accuracy, so as to improve the accuracy rate of monitoring exceptional service;In addition, this specification embodiment realizes
The automation of alert threshold determines that compared with determining alert threshold by manual type, efficiency and accuracy are all improved.
Corresponding to the monitoring method for the abnormal traffic that embodiment corresponding to this specification Fig. 5 provides, it is based on identical thinking,
This specification embodiment additionally provides a kind of monitoring device of abnormal traffic, for executing embodiment corresponding to this specification Fig. 5
Method, Fig. 7 is the module composition schematic diagram for the monitoring device of abnormal traffic that this specification embodiment provides, shown in Fig. 7
Device includes:
Module 702 is obtained, for obtaining the true industry of operation system to be monitored target service handled by the setting moment
Business data, and, prediction model is obtained to the operation system to be monitored target industry handled by the setting moment
The prediction business datum that business is predicted;
Determining module 704, for determining the prediction difference data of the prediction business datum and the actual services data;
Comparison module 706, for the prediction difference data to be compared with predetermined alert threshold, with determination
Whether the target service there is exception at the setting moment;Wherein, the alert threshold is based on the prediction model to each
The prediction data and each sample business that sample business is predicted in the business datum of given time are in the true of given time
Prediction difference data between real business datum are determined.
Optionally, above-mentioned alert threshold is determined in the test phase of prediction model;Above-mentioned each sample business is to prediction
Each test sample that model is tested.
Optionally, mean value and standard deviation institute of the above-mentioned alert threshold based on prediction difference data corresponding to each sample business
Determining.
Optionally, median and median of the above-mentioned alert threshold based on prediction difference data corresponding to each sample business
Determined by absolute deviation.
Optionally, setting percentile institute of the above-mentioned alert threshold based on prediction difference data corresponding to each sample business
Determining.
The monitoring device for the abnormal traffic that this specification embodiment provides can also carry out the monitoring device of abnormal traffic in Fig. 5
The method of execution, and realize the function of the monitoring device of abnormal traffic embodiment shown in Fig. 5, details are not described herein.
The monitoring device for the abnormal traffic that this specification embodiment provides, for abnormal industry in determining service monitoring system
When the alert threshold of business monitoring, the achievement data based on prediction model to the index to be predicted of sample business at the setting moment is carried out
Prediction of the index to be predicted of the prediction data of prediction and the sample business between the truthful data corresponding to the setting moment
Difference data is determined;In this way, the prediction error of the prediction model in service monitoring system is taken into account, improves and determine
Alert threshold accuracy, so as to improve the accuracy rate of monitoring exceptional service;In addition, this specification embodiment realizes
The automation of alert threshold determines that compared with determining alert threshold by manual type, efficiency and accuracy are all improved.
Further, based on method shown in above-mentioned Fig. 1 to Fig. 4, this specification embodiment additionally provides a kind of abnormal industry
The monitoring device of business, as shown in Figure 8.
The monitoring device of abnormal traffic can generate bigger difference because configuration or performance are different, may include one or
More than one processor 801 and memory 802 can store one or more storages in memory 802 using journey
Sequence or data.Wherein, memory 802 can be of short duration storage or persistent storage.The application program for being stored in memory 802 can be with
Including one or more modules (diagram is not shown), each module may include one in the monitoring device to abnormal traffic
Family computer executable instruction information.Further, processor 801 can be set to communicate with memory 802, in exception
The series of computation machine executable instruction information in memory 802 is executed in the monitoring device of business.The monitoring of abnormal traffic is set
Standby can also include one or more power supplys 803, one or more wired or wireless network interfaces 804, one or
More than one input/output interface 805, one or more keyboards 806 etc..
In a specific embodiment, the monitoring device of abnormal traffic include memory and one or one with
On program, perhaps more than one program is stored in memory and one or more than one program can wrap for one of them
Include one or more modules, and each module may include that series of computation machine in monitoring device to abnormal traffic can
Information is executed instruction, and is configured to execute this by one or more than one processor or more than one program includes
For carrying out following computer executable instructions information:
The achievement data using prediction model to the index to be predicted of each sample business at the setting moment is obtained to predict
Prediction data, and, obtain the index to be predicted of each sample business truthful data corresponding to the setting moment;Wherein, in advance
Surveying model is the model in service monitoring system for being predicted the index to be predicted of target service;
According to prediction data and truthful data corresponding to each sample business, pre- error of measurement corresponding to each sample business is determined
Value Data;
Based on prediction difference data corresponding to each sample business, determine that rule determines business monitoring according to the threshold value of setting
System is used for the alert threshold being monitored to abnormal traffic.
Optionally, computer executable instructions information when executed, based on prediction difference corresponding to each sample business
Data determine that rule determines alert threshold corresponding to service monitoring system according to the threshold value of setting, comprising:
Determine the mean value and standard deviation of prediction difference data corresponding to each sample business;
Based on mean value and standard deviation, alert threshold corresponding to service monitoring system is determined.
Optionally, computer executable instructions information when executed, is based on mean value and standard deviation, determines business monitoring system
The corresponding alert threshold of system, comprising:
According to mean value and standard deviation, alert threshold is calculated by following formula;
T=μ+N* σ
Wherein, in above-mentioned formula, T indicates that alert threshold corresponding to service monitoring system, μ indicate mean value, and σ indicates mark
Quasi- poor, N indicates constant.
Optionally, computer executable instructions information when executed, based on prediction difference corresponding to each sample business
Data determine that rule determines that service monitoring system is used for the alert threshold of monitoring exceptional service according to the threshold value of setting, comprising:
Determine the median and median absolute deviation of prediction difference data corresponding to each sample business;
Based on median and median absolute deviation, alert threshold corresponding to service monitoring system is determined.
Optionally, computer executable instructions information when executed, is based on median and median absolute deviation, determines
Alert threshold corresponding to service monitoring system, comprising:
According to median and median absolute deviation, alert threshold is calculated by following formula;
T=m+N*MAD
Wherein, in above-mentioned formula, T indicates that alert threshold corresponding to service monitoring system, m indicate median, MAD table
Show median absolute deviation, N indicates constant.
Optionally, computer executable instructions information is when executed, above-mentioned based on prediction corresponding to each sample business
Difference data determines the police that rule determines that above-mentioned service monitoring system is used to be monitored abnormal traffic according to the threshold value of setting
Report threshold value, comprising:
Calculate the setting percentile of prediction difference data corresponding to each sample business;Above-mentioned setting percentile is true
It is set to alert threshold corresponding to service monitoring system.
Optionally, when executed, above-mentioned prediction model is shot and long term memory network to computer executable instructions information
LSTM prediction model.
The monitoring device for the abnormal traffic that this specification embodiment provides, for abnormal industry in determining service monitoring system
When the alert threshold of business monitoring, the achievement data based on prediction model to the index to be predicted of sample business at the setting moment is carried out
Prediction of the index to be predicted of the prediction data of prediction and the sample business between the truthful data corresponding to the setting moment
Difference data is determined;In this way, the prediction error of the prediction model in service monitoring system is taken into account, improves and determine
Alert threshold accuracy, so as to improve the accuracy rate of monitoring exceptional service;In addition, this specification embodiment realizes
The automation of alert threshold determines that compared with determining alert threshold by manual type, efficiency and accuracy are all improved.
Further, it is based on above-mentioned method shown in fig. 5, this specification embodiment additionally provides a kind of prison of abnormal traffic
Equipment is controlled, the structure of the monitoring device of the abnormal traffic can refer to the monitoring device of abnormal traffic shown in Fig. 8.
In a specific embodiment, the monitoring device of abnormal traffic include memory and one or one with
On program, perhaps more than one program is stored in memory and one or more than one program can wrap for one of them
Include one or more modules, and each module may include that series of computation machine in monitoring device to abnormal traffic can
Information is executed instruction, and is configured to execute this by one or more than one processor or more than one program includes
For carrying out following computer executable instructions information:
The actual services data of operation system to be monitored target service handled by the setting moment are obtained, and, it obtains
Prediction model treats the prediction business datum that monitoring business system target service handled by the setting moment is predicted;
Determine the prediction difference data of prediction business datum and actual services data;
Prediction difference data are compared with predetermined alert threshold, with determine target service setting the moment be
No appearance is abnormal;Wherein, alert threshold predicts each sample business in the business datum of given time based on prediction model
Prediction difference data between the actual services data of given time of prediction data and each sample business determined.
Optionally, computer executable instructions information when executed, test rank of the above-mentioned alert threshold in prediction model
Section determines;
Above-mentioned each sample business is each test sample tested above-mentioned prediction model.
Optionally, when executed, it is right that above-mentioned alert threshold is based on each sample business institute to computer executable instructions information
The mean value and standard deviation for the prediction difference data answered are determined.
Optionally, when executed, it is right that above-mentioned alert threshold is based on each sample business institute to computer executable instructions information
The median and median absolute deviation for the prediction difference data answered are determined.
The monitoring device for the abnormal traffic that this specification embodiment provides is used when being monitored to abnormal traffic
Alert threshold, then be that the achievement data based on prediction model to the operational indicator to be predicted of sample business at the setting moment carries out
The operational indicator to be predicted of the prediction data of prediction and the sample business is between the truthful data corresponding to the setting moment
Prediction difference data are determined;In this way, the prediction error of the prediction model in service monitoring system is taken into account, institute is improved
The accuracy of determining alert threshold, so as to improve the accuracy rate of monitoring exceptional service.
Further, based on method shown in above-mentioned Fig. 1 to Fig. 4, this specification embodiment additionally provides a kind of storage Jie
Matter, for storing computer executable instructions information, in a kind of specific embodiment, the storage medium can for USB flash disk, CD,
Hard disk etc., the computer executable instructions information of storage medium storage are able to achieve following below scheme when being executed by processor:
The achievement data using prediction model to the index to be predicted of each sample business at the setting moment is obtained to predict
Prediction data, and, obtain the index to be predicted of each sample business truthful data corresponding to the setting moment;Wherein, in advance
Surveying model is the model in service monitoring system for being predicted the index to be predicted of target service;
According to prediction data and truthful data corresponding to each sample business, pre- error of measurement corresponding to each sample business is determined
Value Data;
Based on prediction difference data corresponding to each sample business, determine that rule determines business monitoring according to the threshold value of setting
System is used for the alert threshold being monitored to abnormal traffic.
Optionally, the computer executable instructions information of storage medium storage is based on various kinds when being executed by processor
Prediction difference data corresponding to this business determine that rule determines alarm corresponding to service monitoring system according to the threshold value of setting
Threshold value, comprising:
Determine the mean value and standard deviation of prediction difference data corresponding to each sample business;
Based on mean value and standard deviation, alert threshold corresponding to service monitoring system is determined.
Optionally, the computer executable instructions information of storage medium storage is based on mean value when being executed by processor
And standard deviation, determine alert threshold corresponding to service monitoring system, comprising:
According to mean value and standard deviation, alert threshold is calculated by following formula;
T=μ+N* σ
Wherein, in above-mentioned formula, T indicates that alert threshold corresponding to service monitoring system, μ indicate mean value, and σ indicates mark
Quasi- poor, N indicates constant.
Optionally, the computer executable instructions information of storage medium storage is based on various kinds when being executed by processor
Prediction difference data corresponding to this business determine that rule determines that service monitoring system is used for abnormal traffic according to the threshold value of setting
The alert threshold of monitoring, comprising:
Determine the median and median absolute deviation of prediction difference data corresponding to each sample business;
Based on median and median absolute deviation, alert threshold corresponding to service monitoring system is determined.
Optionally, the computer executable instructions information of storage medium storage is based on middle position when being executed by processor
Several and median absolute deviation, determines alert threshold corresponding to service monitoring system, comprising:
According to median and median absolute deviation, alert threshold is calculated by following formula;
T=m+N*MAD
Wherein, in above-mentioned formula, T indicates that alert threshold corresponding to service monitoring system, m indicate median, MAD table
Show median absolute deviation, N indicates constant.
Optionally, the computer executable instructions information of storage medium storage is above-mentioned to be based on when being executed by processor
Prediction difference data corresponding to each sample business determine that rule determines that above-mentioned service monitoring system is used for according to the threshold value of setting
The alert threshold that abnormal traffic is monitored, comprising:
Calculate the setting percentile of prediction difference data corresponding to each sample business;Above-mentioned setting percentile is true
It is set to alert threshold corresponding to service monitoring system.
Optionally, the computer executable instructions information of storage medium storage is when being executed by processor, above-mentioned prediction
Model is shot and long term memory network LSTM prediction model.
The computer executable instructions information for the storage medium storage that this specification embodiment provides is being executed by processor
When, when being used for the alert threshold of monitoring exceptional service in determining service monitoring system, based on prediction model to sample business
The index to be predicted of prediction data and the sample business that achievement data of the index to be predicted at the setting moment is predicted exists
Prediction difference data between truthful data corresponding to the setting moment are determined;In this way, by the prediction in service monitoring system
The prediction error of model is taken into account, and the accuracy of identified alert threshold is improved, so as to improve abnormal traffic prison
The accuracy rate of control;In addition, the automation that this specification embodiment realizes alert threshold determines, warned with being determined by manual type
Report threshold value is compared, and efficiency and accuracy are all improved.
Further, it is based on above-mentioned method shown in fig. 5, this specification embodiment additionally provides a kind of storage medium, uses
In a kind of storage computer executable instructions information, specific embodiment, which can be USB flash disk, CD, hard disk
Computer executable instructions information Deng the storage of, the storage medium is able to achieve following below scheme when being executed by processor:
The actual services data of operation system to be monitored target service handled by the setting moment are obtained, and, it obtains
Prediction model treats the prediction business datum that monitoring business system target service handled by the setting moment is predicted;
Determine the prediction difference data of prediction business datum and actual services data;
Prediction difference data are compared with predetermined alert threshold, with determine target service setting the moment be
No appearance is abnormal;Wherein, alert threshold predicts each sample business in the business datum of given time based on prediction model
Prediction difference data between the actual services data of given time of prediction data and each sample business determined.
Optionally, the computer executable instructions information of storage medium storage is when being executed by processor, above-mentioned alarm
Threshold value is determined in the test phase of prediction model;
Above-mentioned each sample business is each test sample tested above-mentioned prediction model.
Optionally, the storage medium storage computer executable instructions information when being executed by processor, above-mentioned alarm
Threshold value is determined based on the mean value and standard deviation of prediction difference data corresponding to each sample business.
Optionally, the storage medium storage computer executable instructions information when being executed by processor, above-mentioned alarm
Threshold value is determined based on the median and median absolute deviation of prediction difference data corresponding to each sample business.
The computer executable instructions information for the storage medium storage that this specification embodiment provides is being executed by processor
When, when being monitored to abnormal traffic, used alert threshold is then based on prediction model to the to be predicted of sample business
The operational indicator to be predicted of prediction data and the sample business that achievement data of the operational indicator at the setting moment is predicted
Prediction difference data between the truthful data corresponding to the setting moment are determined;In this way, by pre- in service monitoring system
The prediction error for surveying model is taken into account, and the accuracy of identified alert threshold is improved, so as to improve abnormal traffic
The accuracy rate of monitoring.
In the 1990s, the improvement of a technology can be distinguished clearly be on hardware improvement (for example,
Improvement to circuit structures such as diode, transistor, switches) or software on improvement (improvement for method flow).So
And with the development of technology, the improvement of current many method flows can be considered as directly improving for hardware circuit.
Designer nearly all obtains corresponding hardware circuit by the way that improved method flow to be programmed into hardware circuit.Cause
This, it cannot be said that the improvement of a method flow cannot be realized with hardware entities module.For example, programmable logic device
(Programmable Logic Device, PLD) (such as field programmable gate array (Field Programmable Gate
Array, FPGA)) it is exactly such a integrated circuit, logic function determines device programming by user.By designer
Voluntarily programming comes a digital display circuit " integrated " on a piece of PLD, designs and makes without asking chip maker
Dedicated IC chip.Moreover, nowadays, substitution manually makes IC chip, this programming is also used instead mostly " is patrolled
Volume compiler (logic compiler) " software realizes that software compiler used is similar when it writes with program development,
And the source code before compiling also write by handy specific programming language, this is referred to as hardware description language
(Hardware Description Language, HDL), and HDL is also not only a kind of, but there are many kind, such as ABEL
(Advanced Boolean Expression Language)、AHDL(Altera Hardware Description
Language)、Confluence、CUPL(Cornell University Programming Language)、HDCal、JHDL
(Java Hardware Description Language)、Lava、Lola、MyHDL、PALASM、RHDL(Ruby
Hardware Description Language) etc., VHDL (Very-High-Speed is most generally used at present
Integrated Circuit Hardware Description Language) and Verilog.Those skilled in the art also answer
This understands, it is only necessary to method flow slightly programming in logic and is programmed into integrated circuit with above-mentioned several hardware description languages,
The hardware circuit for realizing the logical method process can be readily available.
Controller can be implemented in any suitable manner, for example, controller can take such as microprocessor or processing
The computer for the computer readable program code (such as software or firmware) that device and storage can be executed by (micro-) processor can
Read medium, logic gate, switch, specific integrated circuit (Application Specific Integrated Circuit,
ASIC), the form of programmable logic controller (PLC) and insertion microcontroller, the example of controller includes but is not limited to following microcontroller
Device: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicone Labs C8051F320 are deposited
Memory controller is also implemented as a part of the control logic of memory.It is also known in the art that in addition to
Pure computer readable program code mode is realized other than controller, can be made completely by the way that method and step is carried out programming in logic
Controller is obtained to come in fact in the form of logic gate, switch, specific integrated circuit, programmable logic controller (PLC) and insertion microcontroller etc.
Existing identical function.Therefore this controller is considered a kind of hardware component, and to including for realizing various in it
The device of function can also be considered as the structure in hardware component.Or even, it can will be regarded for realizing the device of various functions
For either the software module of implementation method can be the structure in hardware component again.
System, device, module or the unit that above-described embodiment illustrates can specifically realize by computer chip or entity,
Or it is realized by the product with certain function.It is a kind of typically to realize that equipment is computer.Specifically, computer for example may be used
Think personal computer, laptop computer, cellular phone, camera phone, smart phone, personal digital assistant, media play
It is any in device, navigation equipment, electronic mail equipment, game console, tablet computer, wearable device or these equipment
The combination of equipment.
For convenience of description, it is divided into various units when description apparatus above with function to describe respectively.Certainly, implementing this
The function of each unit can be realized in the same or multiple software and or hardware when application.
It should be understood by those skilled in the art that, embodiments herein can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the application, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The application is reference according to the method for this specification embodiment, the stream of equipment (system) and computer program product
Journey figure and/or block diagram describe.It should be understood that can be by computer program instructions information realization flowchart and/or the block diagram
The combination of process and/or box in each flow and/or block and flowchart and/or the block diagram.It can provide these calculating
Machine program instruction information is to general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices
Processor is to generate a machine, so that the instruction executed by computer or the processor of other programmable data processing devices
Information generates specifies for realizing in one or more flows of the flowchart and/or one or more blocks of the block diagram
Function device.
These computer program instructions information, which may also be stored in, is able to guide computer or other programmable data processing devices
In computer-readable memory operate in a specific manner, so that command information stored in the computer readable memory produces
Raw includes the manufacture of command information device, the command information device realize in one or more flows of the flowchart and/or
The function of being specified in one or more blocks of the block diagram.
These computer program instructions information also can be loaded onto a computer or other programmable data processing device, so that
Series of operation steps are executed on a computer or other programmable device to generate computer implemented processing, thus calculating
The command information that is executed on machine or other programmable devices provide for realizing in one or more flows of the flowchart and/or
The step of function of being specified in one or more blocks of the block diagram.
In a typical configuration, calculating equipment includes one or more processors (CPU), input/output interface, net
Network interface and memory.
Memory may include the non-volatile memory in computer-readable medium, random access memory (RAM) and/or
The forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable medium
Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer-readable instruction information, data structure, the module of program or other numbers
According to.The example of the storage medium of computer includes, but are not limited to phase change memory (PRAM), static random access memory
(SRAM), dynamic random access memory (DRAM), other kinds of random access memory (RAM), read-only memory
(ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory techniques, CD-ROM are read-only
Memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape magnetic disk storage or
Other magnetic storage devices or any other non-transmission medium, can be used for storage can be accessed by a computing device information.According to
Herein defines, and computer-readable medium does not include temporary computer readable media (transitory media), such as modulation
Data-signal and carrier wave.
It should also be noted that, the terms "include", "comprise" or its any other variant are intended to nonexcludability
It include so that the process, method, commodity or the equipment that include a series of elements not only include those elements, but also to wrap
Include other elements that are not explicitly listed, or further include for this process, method, commodity or equipment intrinsic want
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including described want
There is also other identical elements in the process, method of element, commodity or equipment.
It will be understood by those skilled in the art that embodiments herein can provide as method, system or computer program product.
Therefore, complete hardware embodiment, complete software embodiment or embodiment combining software and hardware aspects can be used in the application
Form.It is deposited moreover, the application can be used to can be used in the computer that one or more wherein includes computer usable program code
The shape for the computer program product implemented on storage media (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
Formula.
The application can computer executable instructions information it is general up and down described in the text, such as
Program module.Generally, program module include routines performing specific tasks or implementing specific abstract data types, it is program, right
As, component, data structure etc..The application can also be practiced in a distributed computing environment, in these distributed computing environment
In, by executing task by the connected remote processing devices of communication network.In a distributed computing environment, program module
It can be located in the local and remote computer storage media including storage equipment.
All the embodiments in this specification are described in a progressive manner, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for system reality
For applying example, since it is substantially similar to the method embodiment, so being described relatively simple, related place is referring to embodiment of the method
Part explanation.
The above description is only an example of the present application, is not intended to limit this application.For those skilled in the art
For, various changes and changes are possible in this application.All any modifications made within the spirit and principles of the present application are equal
Replacement, improvement etc., should be included within the scope of the claims of this application.