CN108920336A - A kind of service abnormity prompt method and device based on time series - Google Patents

A kind of service abnormity prompt method and device based on time series Download PDF

Info

Publication number
CN108920336A
CN108920336A CN201810514840.8A CN201810514840A CN108920336A CN 108920336 A CN108920336 A CN 108920336A CN 201810514840 A CN201810514840 A CN 201810514840A CN 108920336 A CN108920336 A CN 108920336A
Authority
CN
China
Prior art keywords
period
time
prediction
default
preset
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201810514840.8A
Other languages
Chinese (zh)
Inventor
苏思洋
李涛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Kylin Seing Network Technology Ltd By Share Ltd
Original Assignee
Kylin Seing Network Technology Ltd By Share Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Kylin Seing Network Technology Ltd By Share Ltd filed Critical Kylin Seing Network Technology Ltd By Share Ltd
Priority to CN201810514840.8A priority Critical patent/CN108920336A/en
Publication of CN108920336A publication Critical patent/CN108920336A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/30Monitoring
    • G06F11/32Monitoring with visual or acoustical indication of the functioning of the machine
    • G06F11/324Display of status information
    • G06F11/327Alarm or error message display
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/30Monitoring
    • G06F11/3003Monitoring arrangements specially adapted to the computing system or computing system component being monitored
    • G06F11/302Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system component is a software system

Abstract

The service abnormity prompt method and device based on time series that this application discloses a kind of, the method includes:The default multiple true values for reminding index within the period of designated length are obtained according to prefixed time interval;The multiple predicted values of the default prompting index during the period of time are determined according to prediction model, and the prediction model is that the time series based on the default history true value for reminding index is fitted;According to multiple predicted values in the multiple true values and the period in the period, whether determination carries out service abnormity prompt during the period of time;After the prompting end cycle where the period, the default true value for reminding index described in the prompting period is added in the default history true value for reminding index, to be updated to the prediction model.The case where prompting efficiency can be improved in this method and device, reduces wrong report, fails to report.

Description

A kind of service abnormity prompt method and device based on time series
Technical field
This application involves field of computer technology more particularly to a kind of service abnormity prompt method based on time series and Device.
Background technique
With the development of computer technology, the service that terminal device can provide a user is more and more, such as personalized Mobile phone operating system, personalized mobile phone desktop etc..For developer of services, setting service monitoring alarm system is to protect One of the effective means of high availability for hindering service is servicing when something goes wrong, if developer can have found at the first time, and Repair to service is the good method stopped loss in time.
The threshold of index is reminded in current service monitoring alarm system, the subjective experience setting for relying primarily on service developers Value, and reminded when the prompting index of service is more than corresponding threshold value, for example, when servicing the mistake occurred at the appointed time It is reminded when accidentally number is more than setting number.
However, not only needing when the threshold value of index is reminded in the subjective experience setting according to developer by a large amount of Practice can just determine relatively reasonable prompting metrics-thresholds, and change when servicing interdependent internal or external environment When, the case where being adjusted again according to new subjective experience to threshold value, this not only inefficiency, failing to report and report by mistake, is also relatively tighter Weight.
Summary of the invention
The embodiment of the present application provides a kind of service abnormity prompt method and device based on time series, reminds effect to improve Rate, and reduce the case where failing to report and reporting by mistake.
In a first aspect, the embodiment of the present application provides a kind of service abnormity prompt method based on time series, the method Including:
The default multiple true values for reminding index within the period of designated length are obtained according to prefixed time interval, it is described Period was located at before current time, and the period includes the current time;
The multiple predicted values of the default prompting index during the period of time, the prediction mould are determined according to prediction model Type is that the time series based on the default history true value for reminding index is fitted;
According to multiple predicted values in the multiple true values and the period in the period, determine in the time Whether service abnormity prompt is carried out in section;
After the prompting end cycle where the period, by the true of default prompting index described in the prompting period Real value is added in the default history true value for reminding index, to be updated to the prediction model.
Second aspect, the embodiment of the present application also provide a kind of service abnormity prompt device based on time series, the dress Set including:
First obtains module, for obtaining default prompting index within the period of designated length according to prefixed time interval Multiple true values, the period is located at before current time, and the period includes the current time;
Prediction module, for determining the multiple predictions of the default prompting index during the period of time according to prediction model Value, the prediction model are that the time series based on the default history true value for reminding index is fitted;
Reminding module, for according to multiple predicted values in the multiple true values and the period in the period, Whether determination carries out service abnormity prompt during the period of time;
Adding module mentions described preset in the prompting period after the prompting end cycle where the period The true value for index of waking up is added in the default history true value for reminding index, to be updated to the prediction model.
At least one above-mentioned technical solution that the embodiment of the present application uses, on the one hand, refer to since default remind can be based on The prediction model that the time series of target history true value is fitted, predict it is default remind index at current time it is previous Predicted value in the section time, and by comparing the true value reminded index during this period of time and predicted value is preset, it automatically determines Whether service abnormity prompt is carried out during this period of time, sentenced without the subjective experience given threshold according to service developers It is disconnected whether to carry out service abnormity prompt, therefore prompting efficiency can be improved;On the other hand, due to that can will be mentioned where current time The true value for the default prompting index in the period of waking up is added in the default history true value for reminding index, to described pre- Model is surveyed to be updated, so that at least one above-mentioned technical solution has ability of self-teaching, it can over time certainly It is dynamic to update prediction model, therefore the case where reporting by mistake and failing to report can be reduced.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present application, constitutes part of this application, this Shen Illustrative embodiments and their description please are not constituted an undue limitation on the present application for explaining the application.In the accompanying drawings:
Fig. 1 is that a kind of process of the service abnormity prompt method provided by the embodiments of the present application based on time series is illustrated Figure.
Fig. 2 is the default a kind of distribution schematic diagram for reminding index provided by the embodiments of the present application.
Fig. 3 is the default another distribution schematic diagram for reminding index provided by the embodiments of the present application.
Fig. 4 is a kind of detailed process schematic diagram of the step S103 in embodiment shown in FIG. 1.
Fig. 5 is that another process of the service abnormity prompt method provided by the embodiments of the present application based on time series is illustrated Figure.
Fig. 6 is a kind of structural representation of the service abnormity prompt device provided by the embodiments of the present application based on time series Figure.
Fig. 7 is a kind of detailed construction schematic diagram of the module 603 in embodiment shown in fig. 6.
Fig. 8 is the structural schematic diagram of a kind of electronic equipment provided by the embodiments of the present application.
Specific embodiment
To keep the purposes, technical schemes and advantages of the application clearer, below in conjunction with the application specific embodiment and Technical scheme is clearly and completely described in corresponding attached drawing.Obviously, described embodiment is only the application one Section Example, instead of all the embodiments.Based on the embodiment in the application, those of ordinary skill in the art are not doing Every other embodiment obtained under the premise of creative work out, shall fall in the protection scope of this application.
In order to solve prompting scheme inefficiency in the prior art, and the serious problem of comparison, this Shen are failed to report in wrong report Please embodiment provide a kind of service abnormity prompt method and apparatus based on time series, be illustrated separately below.
It should be noted that implementing a kind of service abnormity prompt method based on time series provided by the embodiments of the present application And the executing subject of device, it can be terminal and be also possible to server, the embodiment of the present application is to implementing the above method and device Main body is embodied without limitation.
It should also be noted that, the service addressed in the embodiment of the present application is either online service is also possible to take under line Business, online service for example can be cloud storage service, network service, electric business service etc., and servicing under line for example can be terminal Operating system service etc..
First a kind of service abnormity prompt method based on time series provided by the embodiments of the present application is illustrated below.
As shown in Figure 1, a kind of service abnormity prompt method based on time series provided by the embodiments of the present application, can wrap Include following steps:
Step 101 reminds index multiple true within the period of designated length according to prefixed time interval acquisition is default Real value.
The above-mentioned period was located at before current time, and the period includes the current time.Between the above-mentioned pre- time Every the prompting period of usually less than service abnormity prompt, (reminding the determination in period be will be described below in detail, and wouldn't say herein It is bright), and/or, above-mentioned designated length is less than the prompting period.
Default prompting index is to measure the index of destination service operating condition, for example, default remind index to can be use There is abnormal or mistake number in the service of the reporting of user of the service;For another example, it presets and index is reminded to can also be that user accesses Access frequency or access times of the service, etc..
For example, remind index multiple true within the period of designated length according to prefixed time interval acquisition is default Real value can be:It was obtained every 15 minutes in default 15 minutes for reminding index before current time (including current time) Each minute corresponding true value, totally 15.
In the embodiment of the present application, above-mentioned prefixed time interval can be equal with above-mentioned designated length, can also be unequal. For example, prefixed time interval and the designated length all can be 15 minutes or prefixed time interval can be 5 minutes, and The designated length is 15 minutes.
Step 102 determines the multiple predicted values of the default prompting index during the period of time according to prediction model.
Wherein, prediction model is that the time series based on the default history true value for reminding index is fitted. The time series of true value, it can be understood as be the true value of the corresponding distribution of sequence elapsed backward naturally according to the time.Specifically As shown in Figures 2 and 3, presetting reminds the true value of index right according to the sequence elapsed naturally in this way from April 19 to April 25 It should be distributed.
Since user has periodical, seasonality using the behavior of service, so that the default distribution for reminding index Have periodical, seasonality, specific as shown in Fig. 2, default remind index --- the number of mistake appearance has in time series Two apparent features:First is that periodically, specifically the number that occurs of mistake (namely is mentioned daily variation tendency is roughly the same Period of waking up is equal to 1 day), daily at noon with occur noon peak and evening peak respectively at night, and daily noon peak and evening peak Value it is relatively stable, 8 points of morning or so mistake occur numbers fluctuate;Second is that independence, the number that mistake occurs only by The influence of service system internal environment will not change with the fluctuation of flowing of access, i.e. history true value (also referred to as sample) number When according to measuring sufficiently large, what the variation of mistake frequency of occurrence and service internal system occurred anomaly exists one-to-one relationship.Therefore, It can be fitted by the time series to the default history true value for reminding index, obtain to predict the pre- of a certain period If reminding the prediction model for referring to target value.
Further, since Three-exponential Smoothing algorithm can be to simultaneously containing trend and the progress of seasonal time series Prediction, therefore, as an example, a kind of service abnormity prompt method based on time series provided by the embodiments of the present application is also May include:Obtain the time series of the default history true value for reminding index;According to Three-exponential Smoothing algorithm and institute The time series for stating history true value determines the prediction model.
More specifically, it can be determined first according to the time series of Three-exponential Smoothing algorithm and the history true value The default residual error data prediction model for reminding index, the default tendency data prediction model for reminding index and institute State the default Seasonal Data prediction model for reminding index;Then by the residual error data prediction model, the tendency data Prediction model and the cumulative or tired of the Seasonal Data prediction model multiply prediction model, are determined as predicting that the default prompting refers to Target prediction model.Can illustrate introduction below, determine the default prediction model for reminding index using Three-exponential Smoothing algorithm Process.
It is first flat to single exponential smoothing algorithm and secondary index below in order to be more clearly understood that Three-exponential Smoothing algorithm Sliding algorithm is briefly described.
Single exponential smoothing algorithm is based on following recurrence relation:
si=α xi+(1-α)si-1
Wherein, siThe smooth value (referred to as residual error data) of i data before expression current time, α indicate primary smooth system Number, α ∈ [0,1], α are closer to 1, and the smoothed out data value being worth closer to current time, data are more unsmooth, α closer to 0, The smoothed out smooth value being worth closer to preceding i data, data are more smooth, usually by being constantly trying to find the optimum value of α.
The prediction model obtained based on single exponential smoothing algorithm can be:
xi+h=si
Wherein, xi+hIndicate predicted value, i is at the time of the last one history true value corresponds to (corresponding to the horizontal seat in Fig. 2 Mark).It is not difficult to find out that based on the prediction model that single exponential smoothing algorithm is fitted, the time series predicted is Yi Tiaoshui Flat line is unable to the trend and seasonality of reflecting time sequence.
Double smoothing algorithm remains the information of trend, so that the time series predicted may include historical time The variation tendency of sequence.Double smoothing algorithm indicates that smoothed out trend (is known as becoming by one new variable t of addition Gesture data) so that there are following recurrence relations for double smoothing algorithm:
si=α xi+(1-α)(si-1+ti-1)
ti=β (si-si-1)+(1-β)ti-1
Wherein, β indicates that secondary smoothing factor, t indicate that tendency data, the physical meaning of remaining symbol are smoothly calculated with primary It is identical in method.
The prediction model of secondary Smoothness Index algorithm can be:
xi+h=si+hti
Wherein, h indicates the duration between two neighboring time point.
Based on the prediction model that double smoothing algorithm is fitted, the time series predicted is one oblique straight Line.It is not difficult to find out that although the change of time series can be predicted based on the prediction model that double smoothing algorithm is fitted Change trend, but it is unable to the periodicity or Rules of Seasonal Changes of predicted time sequence.
Therefore, as an example, Three-exponential Smoothing algorithm is selected to introduce seasonal parameter in the embodiment of the present application P, fitting obtain prediction model.
There are following recurrence relations for Three-exponential Smoothing algorithm:
si=α (xi-pi-k)+(1-α)(si-1+ti-1)
ti=β (si-si-1)+(1-β)ti-1
pi=γ (xi-si)+(1-γ)pi-k
Wherein, α indicates that a smoothing factor, β indicate that secondary smoothing factor, γ indicate smoothing factor three times;In the application Embodiment due to need the value predicted be it is default remind index, specifically, s, which can indicate described, default reminds the residual of index Difference data prediction model, t can indicate that the default tendency data prediction model for reminding index, p can indicate that described preset mentions The Seasonal Data prediction model for index of waking up, i indicate time point, and k indicates the prompting period, and h indicates two neighboring time point Between duration (also referred to as step-length), mod expression take the remainder oeprator.
On this basis, based on Three-exponential Smoothing algorithm it can be concluded that cumulative and tired multiply two kinds of prediction models:
Wherein, cumulative prediction model can be expressed as:
xi+h=si+hti+pi-k+(h mod k)
It is tired to multiply prediction model and be expressed as:
xi+h=(si+hti)pi-k+(h mod k)
Multiply in prediction model in cumulative prediction model with tired, x indicates the default predicted value for reminding index, remaining symbol Physical meaning see above.
Prediction model employed in above-mentioned steps 102 is also possible to tired multiply prediction mould either cumulative prediction model Type.Certainly, in practical applications, can be selected from the two models according to actual prediction effect one preferably as walk Prediction model in rapid 102.
In the embodiment of the present application, the cumulative prediction model in above-mentioned example and the tired parameter alpha multiplied in prediction model, β and The occurrence of γ is constantly to be fitted to the default history true value for reminding index, and the value of α, β and γ are all Between [0,1].
The process for obtaining preferable α, β and γ value to continuous fitting below is briefly described.
Firstly, the initial value of setting α, β and γ, such as their value are followed successively by:0.2,0.1 and 0.45.
Secondly, choosing the default history true value for reminding index and verification true value.
Specifically as shown in figure 3, can be by default prompting index 31 days 00 March:On 00 to April 3 00:00 in this 3 days Time series was as history true value, by 3 days 00 April:On 00 to April 4 00:00 this 1 day time series is true as verification Value.
Again, the initial value of s, t, p are determined, it is generally the case that s0=x0, t0=x1-x0, using p when cumulative prediction model0 =0, using it is tired multiply prediction model when p0=1, wherein x0It is worth for first in the time series of selected history true value, x1For second value in the time series of selected history true value.
Then, the value for constantly changing α, β and γ, until predicted value and the verification that the prediction model fitted predicts are true Real value can coincide well, for example, until utilizing on March 31 00:On 00 to April 3 00:00 time series in this 3 days is quasi- The prediction model closed out, 3 days 00 calculated April:On 00 to April 4 00:00 this 1 day predicted value and this verification in 1 day is true Real value can coincide well.Finally, the value of α, β and γ when will coincide well remind the pre- of index as default The parameter of model is surveyed, to obtain the default prediction model for reminding index.
It should be noted that when being fitted with time series shown in Fig. 3, the value of k can be that 24 (unit is small When), the value of h can be 1 (unit is hour) namely step-length is 1.
It should also be noted that, in the embodiment of the present application, being fitted to obtain prediction model using Three-exponential Smoothing algorithm Mode be only an example, should not be construed as the restriction to the protection scope of the application, in practical applications, can be with benefit It is fitted to obtain the default prediction for reminding index with time series of other algorithms to the default history true value for reminding index Model.
Step 103, according to multiple predicted values in the multiple true values and the period in the period, determine and exist Whether service abnormity prompt is carried out in the period.
In step 103, mainly to multiple predictions in the multiple true values and the period in the period Value is compared, if true value difference compared with predicted value is larger, (difference is larger to mean that exception occurs in service, default to mention Awake index does not meet expection), it is determined that service abnormity prompt is carried out in the period.
Specifically, the period may include the time point of preset quantity, such as the period is 15 minutes, each Minute at corresponding 1 time point, then include 15 time points in the period.
On this basis, in the first instance, above-mentioned steps 103 may include:
Determine the first dispersion degree and the preset quantity of the time point of the preset quantity corresponding true value Second dispersion degree of time point corresponding predicted value;
If first dispersion degree is greater than or equal to the presupposition multiple of second dispersion degree, it is determined that described Service abnormity prompt is carried out in period.
Wherein, dispersion degree can specifically be indicated with variance (also known as standard deviation), very poor or mean difference.
It is appreciated that illustrating true value when the dispersion degree of true value is more than the dispersion degree presupposition multiple of predicted value There is exception, needs to trigger service abnormity prompt.Wherein, presupposition multiple can be manually set, under normal conditions presupposition multiple Greater than 1, such as presupposition multiple can be equal to 2.
For example, it is assumed that the period is 15 minutes, it include 15 time points, then default prompting index can be determined 15 true values during this period of time, 15 predicted values, this corresponding variance of 15 true values are a, and 15 predicted values are corresponding Variance be b, then service abnormity prompt is carried out when meeting following condition:A is greater than or equal to the presupposition multiple of b.
The dispersion degree of true value and the dispersion degree of predicted value are intuitively shown in Fig. 3, specifically, 3 days 00 April: The fluctuation (dash area in Fig. 3) of true value before 00 is smaller (namely the dispersion degree of true value is smaller), on April 3 00: The fluctuation of the predicted value of part is larger (namely the dispersion degree of predicted value is larger) after 00.
It is conceivable that if multiple true values in the period are bigger compared to predicted value dispersion degree, when illustrating this Big fluctuation has occurred in true value in section, and service is possible to exception occurred, needs and alarm, so that developer repairs Failure stops loss in time.
In second example, above-mentioned steps 103 may include:
Determine the quantity of object time point in the time point of the preset quantity, the object time point is corresponding true It is worth the time point for being greater than or equal to preset threshold with the absolute difference of corresponding predicted value;
If the ratio that the quantity of the object time point accounts for the preset quantity is greater than or equal to preset ratio, it is determined that Service abnormity prompt is carried out during the period of time.
Wherein, preset threshold can also be manually set.In practical applications, a threshold function table y=f (x) can be set, Wherein, x indicates the default actual value for reminding index, and determines above-mentioned preset threshold based on actual value x, and incite somebody to action | x-p | ﹥ f (x) Time point be determined as object time point, p indicates the default predicted value for reminding index herein.Preset ratio can also be taking human as setting It is fixed, such as preset ratio can be equal to 30%.
For example, it is assumed that the period is 15 minutes, it include 15 time points, then default prompting index can be determined 15 true values during this period of time, 15 predicted values, the absolute value of 15 differences and 15 differences absolute value in it is big In or equal to preset threshold ratio be c, then service abnormity prompt is carried out when meeting following condition:C is greater than preset ratio.
In third example, as shown in figure 4, above-mentioned steps 103 may include:
Step 401, the first dispersion degree for determining the time point of the preset quantity corresponding true value and described pre- If the second dispersion degree of the time point of quantity corresponding predicted value.
Step 402, the quantity for determining object time point in the time point of the preset quantity, the object time point are pair The true value answered is greater than or equal to the time point of preset threshold with the absolute difference of corresponding predicted value.
If step 403, first dispersion degree are greater than or equal to the presupposition multiple of second dispersion degree, and institute The quantity for stating object time point accounts for the ratio of the preset quantity more than or equal to preset ratio, it is determined that during the period of time The service of progress abnormity prompt.
Wherein, presupposition multiple can be identical as the presupposition multiple in first example, and preset ratio can be with second example Preset ratio in son is identical.
For example, it is assumed that the period is 15 minutes, it include 15 time points, then default prompting index can be determined The absolute value of 15 true values during this period of time, 15 predicted values, 15 differences;Assuming that the corresponding side of this 15 true values Difference is a, and the corresponding variance of 15 predicted values is to be greater than or equal to the ratio of preset threshold in the absolute value of b and 15 difference For c, then service abnormity prompt is carried out when meeting two following conditions at the same time:1, a is greater than or equal to the presupposition multiple of b;2, c is big In preset ratio.
Due in third example by the difference of the dispersion degree and true value of true value and predicted value and predicted value The accounting that absolute value is greater than or equal to preset threshold is combined together, it is determined whether carries out service abnormity prompt, therefore, Ke Yigeng It avoids reporting by mistake well and fail to report.
It should be noted that according to multiple predicted values in the multiple true values and the period in the period, Determine that the mode for whether carrying out service abnormity prompt during the period of time can be not limited to above-mentioned three kinds, in practical applications, Conceived based on present invention, more examples can be expanded, the embodiment of the present application does not limit this.
Step 104, after the prompting end cycle where the period, by the default prompting in the prompting period The true value of index is added in the default history true value for reminding index, to be updated to the prediction model.
As shown in Figures 2 and 3, the period is reminded to can be one day.If specific as shown in Fig. 2, the period is in April 21 Day 00:On 00 to April 22 00:It, then can be 22 days 00 in April in 00 this prompting period:The a certain moment after 00, by April 21 days 00:On 00 to April 22 00:The default true value for reminding index is added to described preset and mentions in 00 this prompting period In the history true value for index of waking up.
A kind of service abnormity prompt method based on time series that embodiment shown in FIG. 1 provides, on the one hand, due to can With the prediction model that the time series based on the default history true value for reminding index is fitted, default prompting index is predicted Current time for the previous period in predicted value, and by comparing it is default remind the true value of index during this period of time with Predicted value automatically determines and whether carries out service abnormity prompt during this period of time, without the master according to service developers Sight experience given threshold judges whether to service abnormity prompt, therefore prompting efficiency can be improved;On the other hand, due to can The true value for the default prompting index reminded in the period where current time is added to the default history for reminding index In true value, the prediction model is updated, so that the above method has ability of self-teaching, it being capable of pushing away with the time Shifting automatically updates prediction model, therefore can reduce the case where reporting by mistake and failing to report.
It is further to a kind of service abnormity prompt method progress based on time series provided by the present application below with reference to Fig. 5 Ground explanation.
As shown in figure 5, a kind of service abnormity prompt method based on time series provided by the embodiments of the present application can wrap It includes:
Step 501 obtains the default multiple true values for reminding index within the period of designated length, and described in determination It is default to remind the multiple predicted values of index during the period of time.
In practical applications, the module for executing step 501 can be known as fallout predictor.
More specifically, step 501 includes:
Step 5015, from the default history true value storing data library 5011 for reminding index, obtain the default prompting The time series of the history true value of index, and be fitted based on the time series of the history true value of acquisition and determine prediction model;
Step 5016, based on acquisition it is described it is default remind index history true value time series, determine remind week Issue is according to 5012;
Step 5017 obtains and described default the multiple true values of index during the period of time is reminded (namely to obtain real-time Data), and determine based on the prediction model determined in step 5015 multiple predicted values in the period.
Step 502, according to multiple predicted values in the multiple true values and the period in the period, determine and exist Whether service abnormity prompt is carried out in the period;
In practical applications, the module for executing step 502 can be known as comparator.
More specifically, step 502 may include:
Step 5021 determines whether the dispersion degree of multiple true values in the period is greater than or equal to the period The presupposition multiple of the dispersion degree of interior multiple predicted values, if so, executing step 5022;Otherwise, terminate this process;
Step 5022 determines in the period in multiple absolute differences of multiple true values and multiple predicted values, surpasses Cross the quantity of preset threshold;Judgement is more than that the quantity of the absolute difference of preset threshold accounts for the ratio of the multiple absolute difference Whether preset ratio is greater than or equal to, if so, thening follow the steps 503;Otherwise, terminate this process.
Step 503, the result that step 502 obtains be when, carry out service abnormity prompt.
In practical applications, the concrete mode for servicing abnormity prompt can be triggering alarm equipment alarm.
Likewise, a kind of service abnormity prompt method based on time series shown in fig. 5, since default mention can be based on The prediction model that is fitted of history true value for index of waking up, predict it is default remind index current time for the previous period Interior predicted value, and by comparing the true value reminded index during this period of time and predicted value is preset, it automatically determines at this Between whether carry out service abnormity prompt in section, judge whether without the subjective experience given threshold according to service developers The service of progress abnormity prompt, therefore prompting efficiency can be improved.
Corresponding to above method embodiment, the embodiment of the present application also provides a kind of services based on time series to mention extremely Awake device, is described below.
As shown in fig. 6, a kind of service abnormity prompt device based on time series provided by the embodiments of the present application, can wrap It includes:First obtains module 601, prediction module 602, reminding module 603 and adding module 604.
First obtains module 601, for obtaining the default time for reminding index in designated length according to prefixed time interval Multiple true values in section, the period were located at before current time, and the period includes the current time.
Prefixed time interval and/or designated length, which are less than, reminds the period.
Default prompting index is to measure the index of destination service operating condition.
In the embodiment of the present application, above-mentioned prefixed time interval can be equal with above-mentioned designated length, can also be unequal.
Prediction module 602 described default reminds index during the period of time multiple for determining according to prediction model Predicted value, the prediction model are that the time series based on the default history true value for reminding index is fitted.
Since user has periodical, seasonality using the behavior of service, so that the default distribution for reminding index Have periodical, seasonality, therefore, can be carried out by the history true value to the default prompting index being distributed in chronological order Fitting, obtains to predict that the default prompting of a certain period refers to the prediction model of target value.
Further, since Three-exponential Smoothing algorithm can be to simultaneously containing trend and the progress of seasonal time series Prediction, therefore, optionally, as an example, device shown in fig. 6 can also include:Second obtains module and prediction model Determining module.
Second obtains module, for obtaining the time series of the default history true value for reminding index.
Prediction model determining module, for the time series according to Three-exponential Smoothing algorithm and the history true value, Determine the prediction model.
In one example, the prediction model determining module is specifically used for:
According to the time series of Three-exponential Smoothing algorithm and the history true value, the default prompting index is determined Residual error data prediction model, the default tendency data prediction model for reminding index and the default season for reminding index Section property data prediction model;
By the residual error data prediction model, the tendency data prediction model and the Seasonal Data prediction model It is cumulative or tired multiply prediction model, be determined as predicting the default prediction model for reminding index.
In one more specifically example, the prediction model can for the residual error data prediction model, it is described become The cumulative prediction model of gesture data prediction model and the Seasonal Data prediction model can specifically be indicated with following formula:
xi+h=(si+hti)pi-k+(h mod k)
Alternatively, in another more specifically example, the prediction model can for the residual error data prediction model, The tendency data prediction model and the tired of the Seasonal Data prediction model multiply prediction model, can specifically use following formula table Show:
xi+h=si+hti+pi-k+(h mod k)
Multiply in prediction model in above-mentioned cumulative prediction model with tired, x indicates the default predicted value for reminding index, s table Show that the default residual error data prediction model for reminding index, t indicate the default tendency data prediction mould for reminding index Type, p indicate that the default Seasonal Data prediction model for reminding index, i indicate time point, and k indicates the prompting period, h Indicate the duration between two neighboring time point, mod expression takes the remainder oeprator.
Wherein, presetting reminds the residual error data prediction model of index to be specifically as follows:
si=α (xi-pi-k)+(1-α)(si-1+ti-1)
The default tendency data prediction model for reminding index is specifically as follows:
ti=β (si-si-1)+(1-β)ti-1
The default Seasonal Data prediction model for reminding index can be:
pi=γ (xi-si)+(1-γ)pi-k
α indicates that a smoothing factor, β indicate that secondary smoothing factor, γ indicate smoothing factor, and the tool of α, β and γ three times Body value is constantly to be fitted to the default history true value for reminding index.
In practical applications, can according to actual prediction effect from cumulative prediction model and it is tired multiply in prediction model select One prediction model preferably as the default prompting index in module 602.And cumulative prediction model multiplies prediction model with tired In parameter alpha, β and γ occurrence, be constantly to it is described it is default remind index history true value be fitted, α, The value of β and γ is all between [0,1].
It should be noted that in the embodiment of the present application, being fitted to obtain prediction model using Three-exponential Smoothing algorithm Mode is only an example, should not be construed as the restriction to the protection scope of the application, in practical applications, can also be utilized Other algorithms are fitted to obtain prediction model to the default history true value for reminding index.
Reminding module 603, for according to multiple predictions in the multiple true values and the period in the period Whether value, determination carry out service abnormity prompt during the period of time.
In reminding module 603, mainly to multiple in the multiple true values and the period in the period Predicted value is compared, if true value difference compared with predicted value is larger, (difference is larger to mean that exception occurs in service, in advance If index is reminded not meet expection), it is determined that service abnormity prompt is carried out in the period.
Specifically, the period may include the time point of preset quantity.
On this basis, in one example, reminding module 603 may include:First determines that submodule and first reminds son Module.Wherein:
First determines submodule, the first discrete journey of the time point corresponding true value for determining the preset quantity Second dispersion degree of the time point corresponding predicted value of degree and the preset quantity;
First reminds submodule, if being greater than or equal to the pre- of second dispersion degree for first dispersion degree If multiple, it is determined that carry out service abnormity prompt during the period of time.
It is conceivable that if multiple true values in the period are bigger compared to predicted value dispersion degree, when illustrating this Big fluctuation has occurred in true value in section, and service is possible to exception occurred, needs and alarm, so that developer repairs Failure stops loss in time.
Optionally, in another example, reminding module 603 may include:Second determines that submodule and second reminds submodule Block, wherein:
Second determines submodule, the quantity of object time point, the mesh in the time point for determining the preset quantity Marking time point is the time point that corresponding true value is greater than or equal to preset threshold with the absolute difference of corresponding predicted value;
Second remind submodule, if the ratio that the quantity for the object time point accounts for the preset quantity be greater than or Equal to preset ratio, it is determined that carry out service abnormity prompt during the period of time.
It is also understood that if in multiple time points in the period being more than the time point corresponding difference of preset ratio Being worth absolute value is more than preset threshold, illustrates that the true value in the period deviates from normal variation tendency, service is possible to occur Exception, needs and alarm, in time stops loss so that developer repairs failure.
Optionally, in another example, as shown in fig. 7, reminding module 603 may include:First determining submodule 701, Second determines that submodule 702 and third remind submodule 703.
First determines submodule 701, and first of the time point corresponding true value for determining the preset quantity is discrete Second dispersion degree of the time point of degree and the preset quantity corresponding predicted value;
Second determines submodule 702, and the quantity of object time point, described in the time point for determining the preset quantity Object time point is the time point that corresponding true value is greater than or equal to preset threshold with the absolute difference of corresponding predicted value;
Third reminds submodule 703, if being greater than or equal to second dispersion degree for first dispersion degree Presupposition multiple, and the quantity of object time point account for the preset quantity ratio be greater than or equal to preset ratio, then really It is fixed to carry out service abnormity prompt during the period of time.
In the example shown in Fig. 7, due to by the dispersion degree of true value and predicted value and true value and predicted value The accounting that the absolute value of difference is greater than or equal to preset threshold is combined together, it is determined whether service abnormity prompt is carried out, therefore, It preferably can avoid reporting by mistake and fail to report.
It should be noted that according to multiple predicted values in the multiple true values and the period in the period, Determine that the mode for whether carrying out service abnormity prompt during the period of time can be not limited to above-mentioned three kinds, in practical applications, Conceived based on present invention, more examples can be expanded, the embodiment of the present application does not limit this.
Adding module 604 will be described default in the prompting period after the prompting end cycle where the period The true value of index is reminded to be added in the default history true value for reminding index, to carry out more to the prediction model Newly.
A kind of service abnormity prompt device based on time series that embodiment shown in fig. 6 provides, on the one hand, due to can With the prediction model that the time series based on the default history true value for reminding index is fitted, default prompting index is predicted Current time for the previous period in predicted value, and by comparing it is default remind the true value of index during this period of time with Predicted value automatically determines and whether carries out service abnormity prompt during this period of time, without the master according to service developers Sight experience given threshold judges whether to service abnormity prompt, therefore prompting efficiency can be improved;On the other hand, due to can The true value for the default prompting index reminded in the period where current time is added to the default history for reminding index In true value, the prediction model is updated, so that above-mentioned apparatus has ability of self-teaching, it being capable of pushing away with the time Shifting automatically updates prediction model, therefore can reduce the case where reporting by mistake and failing to report.
It should be noted that since the content and method embodiment that Installation practice executes is similar, herein to device Embodiment part describes more simple, and related place refers to embodiment of the method part.
Fig. 8 show be a kind of electronic equipment provided by the embodiments of the present application structural schematic diagram.Referring to FIG. 8, hard Part level, the electronic equipment include processor, optionally further comprising internal bus, network interface, memory.Wherein, memory It may include memory, such as high-speed random access memory (Random-Access Memory, RAM), it is also possible to further include non- Volatile memory (non-volatile memory), for example, at least 1 magnetic disk storage etc..Certainly, which may be used also It can include hardware required for other business.
Processor, network interface and memory can be connected with each other by internal bus, which can be ISA (Industry Standard Architecture, industry standard architecture) bus, PCI (Peripheral Component Interconnect, Peripheral Component Interconnect standard) bus or EISA (Extended Industry Standard Architecture, expanding the industrial standard structure) bus etc..The bus can be divided into address bus, data/address bus, control always Line etc..Only to be indicated with a four-headed arrow in Fig. 8, it is not intended that an only bus or a type of convenient for indicating Bus.
Memory, for storing program.Specifically, program may include program code, and said program code includes calculating Machine operational order.Memory may include memory and nonvolatile memory, and provide instruction and data to processor.
Processor is from the then operation into memory of corresponding computer program is read in nonvolatile memory, in logical layer The service abnormity prompt device based on time series is formed on face.Processor executes the program that memory is stored, and specifically uses In the execution service abnormity prompt method provided by the embodiments of the present application based on time series.
The side that the service abnormity prompt device based on time series disclosed in the above-mentioned embodiment illustrated in fig. 6 such as the application executes Method can be applied in processor, or be realized by processor.Processor may be a kind of IC chip, with signal Processing capacity.During realization, each step of the above method can by the integrated logic circuit of the hardware in processor or The instruction of person's software form is completed.Above-mentioned processor can be general processor, including central processing unit (Central Processing Unit, CPU), network processing unit (Network Processor, NP) etc.;It can also be Digital Signal Processing Device (Digital Signal Processor, DSP), specific integrated circuit (Application Specific Integrated Circuit, ASIC), field programmable gate array (Field-Programmable Gate Array, FPGA) or other can Programmed logic device, discrete gate or transistor logic, discrete hardware components.It may be implemented or execute the application implementation Disclosed each method, step and logic diagram in example.General processor can be microprocessor or the processor can also be with It is any conventional processor etc..The step of method in conjunction with disclosed in the embodiment of the present application, can be embodied directly in hardware decoding Processor executes completion, or in decoding processor hardware and software module combination execute completion.Software module can position In random access memory, flash memory, read-only memory, programmable read only memory or electrically erasable programmable memory, register In the storage medium of equal this fields maturation.The storage medium is located at memory, and processor reads the information in memory, in conjunction with it Hardware completes the step of above method.
The embodiment of the present application also proposed a kind of computer readable storage medium, the computer-readable recording medium storage one A or multiple programs, the one or more program include instruction, which holds when by the electronic equipment including multiple application programs When row, the electronic equipment can be made to execute what the service abnormity prompt device in embodiment illustrated in fig. 6 based on time series executed Method, and be specifically used for executing the service abnormity prompt method provided by the embodiments of the present application based on time series.
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 referring to method, the process of equipment (system) and computer program product according to the embodiment of the present application Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates, Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one The step of function of being specified in a box or multiple boxes.
It is described it should be noted that each embodiment in the application is all made of relevant mode, between each embodiment Same and similar part may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially Its, for device embodiment, since it is substantially similar to the method embodiment, so being described relatively simple, related place Illustrate referring to the part of embodiment of the method.
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 element There is also other identical elements in process, method, commodity or equipment.
The above is only embodiments herein, are not intended to limit this application.To those skilled in the art, Various changes and changes are possible in this application.It is all within the spirit and principles of the present application made by any modification, equivalent replacement, Improve etc., it should be included within the scope of the claims of this application.

Claims (14)

1. a kind of service abnormity prompt method based on time series, which is characterized in that the method includes:
The default multiple true values for reminding index within the period of designated length, the time are obtained according to prefixed time interval Section was located at before current time, and the period includes the current time;
According to prediction model determine it is described it is default remind index multiple predicted values during the period of time, the prediction model, It is that the time series based on the default history true value for reminding index is fitted;
According to multiple predicted values in the multiple true values and the period in the period, determine during the period of time Whether service abnormity prompt is carried out;
After the prompting end cycle where the period, by the default true value for reminding index described in the prompting period It is added in the default history true value for reminding index, to be updated to the prediction model.
2. the method according to claim 1, wherein the period include preset quantity time point, and
Multiple predicted values in multiple true values and the period according in the period were determined in the time Service abnormity prompt whether is carried out in section, including:
Determine the first dispersion degree of the time point of the preset quantity corresponding true value and the time of the preset quantity Second dispersion degree of the corresponding predicted value of point;
If first dispersion degree is greater than or equal to the presupposition multiple of second dispersion degree, it is determined that in the time Service abnormity prompt is carried out in section.
3. the method according to claim 1, wherein the period include preset quantity time point, and
Multiple predicted values in multiple true values and the period according in the period were determined in the time Service abnormity prompt whether is carried out in section, including:
Determine the quantity of object time point in the time point of the preset quantity, the object time point be corresponding true value with The absolute difference of corresponding predicted value is greater than or equal to the time point of preset threshold;
If the ratio that the quantity of the object time point accounts for the preset quantity is greater than or equal to preset ratio, it is determined that in institute It states and carries out service abnormity prompt in the period.
4. the method according to claim 1, wherein the period include preset quantity time point, and
Multiple predicted values in multiple true values and the period according in the period were determined in the time Service abnormity prompt whether is carried out in section, including:
Determine the first dispersion degree of the time point of the preset quantity corresponding true value and the time of the preset quantity Second dispersion degree of the corresponding predicted value of point;
Determine the quantity of object time point in the time point of the preset quantity, the object time point be corresponding true value with The absolute difference of corresponding predicted value is greater than or equal to the time point of preset threshold;
If first dispersion degree is greater than or equal to the presupposition multiple of second dispersion degree, and object time point Quantity account for the ratio of the preset quantity and be greater than or equal to preset ratio, it is determined that it is abnormal to carry out service during the period of time It reminds.
5. method according to claim 1-4, which is characterized in that
The prefixed time interval and/or the designated length are less than the prompting period.
6. method according to claim 1-4, which is characterized in that it is described determined according to prediction model it is described pre- If before reminding the multiple predicted values of index during the period of time, the method also includes:
Obtain the time series of the default history true value for reminding index;
According to the time series of Three-exponential Smoothing algorithm and the history true value, the prediction model is determined.
7. according to the method described in claim 6, it is characterized in that, described true according to Three-exponential Smoothing algorithm and the history The time series of real value determines the prediction model, including:
According to the time series of Three-exponential Smoothing algorithm and the history true value, the default residual error for reminding index is determined Data prediction model, the default tendency data prediction model for reminding index and the default seasonality for reminding index Data prediction model;
By the tired of the residual error data prediction model, the tendency data prediction model and the Seasonal Data prediction model Add or tire out and multiply prediction model, is determined as predicting the default prediction model for reminding index.
8. a kind of service abnormity prompt device based on time series, which is characterized in that described device includes:
First obtains module, more within the period of designated length for obtaining default prompting index according to prefixed time interval A true value, the period were located at before current time, and the period includes the current time;
Prediction module, for determining the multiple predicted values of the default prompting index during the period of time according to prediction model, The prediction model is that the time series based on the default history true value for reminding index is fitted;
Reminding module, for determining according to multiple predicted values in the multiple true values and the period in the period Whether carry out service abnormity prompt during the period of time;
Adding module refers to default prompting described in the prompting period after the prompting end cycle where the period Target true value is added in the default history true value for reminding index, to be updated to the prediction model.
9. device according to claim 8, which is characterized in that the period includes the time point of preset quantity, described Reminding module includes:
First determining submodule, the first dispersion degree of the time point corresponding true value for determining the preset quantity, with And the second dispersion degree of the time point corresponding predicted value of the preset quantity;
First reminds submodule, if being greater than or equal to default times of second dispersion degree for first dispersion degree Number, it is determined that carry out service abnormity prompt during the period of time.
10. device according to claim 8, which is characterized in that the period includes the time point of preset quantity, and The reminding module includes:
Second determines submodule, the quantity of object time point in the time point for determining the preset quantity, when the target Between point be the time point of corresponding true value and the absolute difference of corresponding predicted value more than or equal to preset threshold;
Second reminds submodule, if the ratio that the quantity for the object time point accounts for the preset quantity is greater than or equal to Preset ratio, it is determined that carry out service abnormity prompt during the period of time.
11. device according to claim 8, which is characterized in that it is characterized in that, the period includes preset quantity Time point, the reminding module include:
First determining submodule, the first dispersion degree of the time point corresponding true value for determining the preset quantity, with And the second dispersion degree of the time point corresponding predicted value of the preset quantity;
Second determines submodule, the quantity of object time point in the time point for determining the preset quantity, when the target Between point be the time point of corresponding true value and the absolute difference of corresponding predicted value more than or equal to preset threshold;
Third reminds submodule, if being greater than or equal to default times of second dispersion degree for first dispersion degree Number, and the quantity of object time point accounts for the ratio of the preset quantity more than or equal to preset ratio, it is determined that described Service abnormity prompt is carried out in period.
12. according to the described in any item devices of claim 8-11, which is characterized in that
The prefixed time interval and/or the designated length are less than the prompting period.
13. according to the described in any item devices of claim 8-11, which is characterized in that described device further includes:
Second obtains module, for obtaining the time series of the default history true value for reminding index;
Prediction model determining module is determined for the time series according to Three-exponential Smoothing algorithm and the history true value The prediction model.
14. device according to claim 13, which is characterized in that the prediction model determining module is specifically used for:
According to the time series of Three-exponential Smoothing algorithm and the history true value, the default residual error for reminding index is determined Data prediction model, the default tendency data prediction model for reminding index and the default seasonality for reminding index Data prediction model;
By the tired of the residual error data prediction model, the tendency data prediction model and the Seasonal Data prediction model Add or tire out and multiply prediction model, is determined as predicting the default prediction model for reminding index.
CN201810514840.8A 2018-05-25 2018-05-25 A kind of service abnormity prompt method and device based on time series Pending CN108920336A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810514840.8A CN108920336A (en) 2018-05-25 2018-05-25 A kind of service abnormity prompt method and device based on time series

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810514840.8A CN108920336A (en) 2018-05-25 2018-05-25 A kind of service abnormity prompt method and device based on time series

Publications (1)

Publication Number Publication Date
CN108920336A true CN108920336A (en) 2018-11-30

Family

ID=64410089

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810514840.8A Pending CN108920336A (en) 2018-05-25 2018-05-25 A kind of service abnormity prompt method and device based on time series

Country Status (1)

Country Link
CN (1) CN108920336A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110535864A (en) * 2019-08-30 2019-12-03 北京达佳互联信息技术有限公司 Service method for detecting abnormality, device, equipment and storage medium
CN112445682A (en) * 2021-02-01 2021-03-05 连连(杭州)信息技术有限公司 System monitoring method, device, equipment and storage medium

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104181883A (en) * 2014-08-08 2014-12-03 青岛高校信息产业有限公司 Method for processing abnormal data of real-time data acquisition system in real time
CN107086944A (en) * 2017-06-22 2017-08-22 北京奇艺世纪科技有限公司 A kind of method for detecting abnormality and device
US20180060127A1 (en) * 2016-08-24 2018-03-01 Ca, Inc. Reservation of hardware resources in a computer system based on utilization measurements during time ranges

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104181883A (en) * 2014-08-08 2014-12-03 青岛高校信息产业有限公司 Method for processing abnormal data of real-time data acquisition system in real time
US20180060127A1 (en) * 2016-08-24 2018-03-01 Ca, Inc. Reservation of hardware resources in a computer system based on utilization measurements during time ranges
CN107086944A (en) * 2017-06-22 2017-08-22 北京奇艺世纪科技有限公司 A kind of method for detecting abnormality and device

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110535864A (en) * 2019-08-30 2019-12-03 北京达佳互联信息技术有限公司 Service method for detecting abnormality, device, equipment and storage medium
CN112445682A (en) * 2021-02-01 2021-03-05 连连(杭州)信息技术有限公司 System monitoring method, device, equipment and storage medium
CN112445682B (en) * 2021-02-01 2021-05-11 连连(杭州)信息技术有限公司 System monitoring method, device, equipment and storage medium

Similar Documents

Publication Publication Date Title
CN106164864B (en) Predictive load scaling of services
CN108920336A (en) A kind of service abnormity prompt method and device based on time series
US20120095608A1 (en) Demand prediction apparatus, and computer readable, non-transitory storage medium
JP4800299B2 (en) Cost information management system, cost information management method, and cost information management program
JP2004171249A (en) Backup execution decision method for database
US9235665B2 (en) System, method and apparatus for handling power limit restrictions in flash memory devices
JP6374702B2 (en) Energy saving effect calculation method and apparatus
US20150149130A1 (en) Power consumption prediction apparatus, method, and non-transitory computer readable storage medium thereof
US8463616B2 (en) Scheduling a process according to time-varying input prices
JP2012141980A (en) Systems and methods for use in correcting predicted failure in production process
JP2007018322A (en) System, method, and program for estimating amount of energy saved
CN106155699B (en) A kind of management method and mobile terminal of background process
CN105120088A (en) A prompting method and an electronic device
WO2020078093A1 (en) Door lock control method and apparatus, and control device
CN108154252A (en) Method and apparatus for estimating the flow deadline
CN108826572A (en) A kind of air-conditioning state determines method and device
Kanagasabai et al. Ec2bargainhunter: It's easy to hunt for cost savings on amazon ec2!
JP2020022237A (en) Power storage system and control method
JP2012155385A (en) Energy consumption type determination system
US9582765B2 (en) Apparatus, method, and non-transitory computer readable storage medium thereof for recommending an electricity consumption behavior
JP2007087121A (en) Total management apparatus, database system, database management method, and database management program
JP2007310704A (en) Method of estimating service use change
US20210044237A1 (en) Method and apparatus for adaptive control of motor, and storage medium
CN111932042B (en) Adaptive scheduling method, apparatus, and computer storage medium
CN108228414B (en) Linux-based processor process parameter monitoring method and monitoring system

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