CN107480028A - The acquisition methods and device of residual time length workable for disk - Google Patents
The acquisition methods and device of residual time length workable for disk Download PDFInfo
- Publication number
- CN107480028A CN107480028A CN201710601830.3A CN201710601830A CN107480028A CN 107480028 A CN107480028 A CN 107480028A CN 201710601830 A CN201710601830 A CN 201710601830A CN 107480028 A CN107480028 A CN 107480028A
- Authority
- CN
- China
- Prior art keywords
- prediction
- disk
- model
- residual capacity
- forecast model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3051—Monitoring arrangements for monitoring the configuration of the computing system or of the computing system component, e.g. monitoring the presence of processing resources, peripherals, I/O links, software programs
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3409—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3447—Performance evaluation by modeling
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Computing Systems (AREA)
- Debugging And Monitoring (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The Forecasting Methodology and device of remaining number of days workable for the present invention proposes a kind of disk, wherein, method includes:By the multidimensional prediction data input of disk into target prediction model, the residual capacity of disk is predicted;Prediction step when being predicted according to residual capacity adjustment next time;Target prediction model is updated according to prediction step;Multidimensional prediction data are re-entered into the target prediction model after renewal, predict residual capacity, and return to execution and prediction step is adjusted according to residual capacity, and target prediction model is updated according to prediction step, untill residual capacity that the target prediction model prediction after renewal goes out is zero;The prediction step of corresponding target prediction model is as residual time length workable for disk when using residual capacity being zero.Because prediction process is by the way of machine learning, and it with reference to influence the multidimensional monitoring data of disk residual capacity, so as to improve the accuracy of prediction well.
Description
Technical field
The present invention relates to the acquisition methods and dress of residual time length workable for computer realm, more particularly to a kind of disk
Put.
Background technology
At present, the residual capacity of the disk on information technoloy equipment can be monitored, is then based on the residual capacity of adjacent two days, is calculated
Go out intraday disk and use increment.Therefore, the residual capacity that disk is daily in a period of time can be gathered, is then calculated more
Individual intraday disk uses increment.After multiple disks are got using increment, it can calculate average daily
Disk uses increment.And then increment can be used based on the residual capacity on the day of disk and disk average daily, just
Remaining number of days workable for disk can be predicted.
The Forecasting Methodology of above-mentioned remaining number of days is based only upon the residual capacity of disk this monitoring data and is predicted, and does not consider
Other reference factors so that prediction result accuracy is relatively low.
The content of the invention
It is contemplated that at least solves one of technical problem in correlation technique to a certain extent.
Therefore, first purpose of the present invention is a kind of acquisition methods of residual time length workable for proposing disk, lead to
Cross machine learning mode and build prediction module, the monitoring data of multidimensional is learnt based on forecast model, to predict that disk can
The residual time length used, the purpose of residual time length accuracy workable for improving disk is realized, to solve in the prior art only base
Residual time length, which is predicted, in the residual capacity of disk this monitoring data the problem of forecasting accuracy is relatively low be present.
Second object of the present invention is a kind of acquisition device of residual time length workable for proposing disk.
Third object of the present invention is to propose a kind of computer equipment.
Fourth object of the present invention is to propose a kind of computer program product.
The 5th purpose of the present invention is to propose a kind of non-transitorycomputer readable storage medium.
For the above-mentioned purpose, the acquisition of residual time length workable for first aspect present invention embodiment proposes a kind of disk
Method, including:
By the multidimensional prediction data input of disk into target prediction model, the residual capacity of the disk is predicted;
Prediction step when being predicted according to residual capacity adjustment next time;
Target prediction model is updated according to the prediction step;
The multidimensional prediction data are re-entered into the target prediction model after renewal, predict the residue
Capacity, and return to execution and the prediction step is adjusted according to the residual capacity, and according to the prediction step to the mesh
Mark forecast model is updated, untill the residual capacity that the target prediction model prediction after renewal goes out is zero;
The prediction step of the corresponding target prediction model can as the disk when using the residual capacity being zero
The residual time length used.
A kind of possible implementation provided as first aspect present invention embodiment, it is described according to the prediction step
The target prediction model is updated, including:
The first forecast model corresponding with the prediction step is obtained according to the prediction step;
It is first forecast model by the target prediction model modification.
A kind of possible implementation provided as first aspect present invention embodiment, it is described according to the prediction step
Before obtaining target prediction model corresponding with the prediction step, in addition to:
The multidimensional history monitoring data of disk is gathered as training data;
In advance according to different prediction steps, the forecast model of structure is trained using the training data, obtained
The first forecast model corresponding with each prediction step.
A kind of possible implementation provided as first aspect present invention embodiment, it is described according to the prediction step
The target prediction model is updated, including:
Using the multidimensional history monitoring data of the disk as training data, it is re-entered into the forecast model of structure,
The forecast model is trained according to the prediction step, obtains the first forecast model corresponding with the prediction step;
It is first forecast model by the target prediction model modification.
A kind of possible implementation provided as first aspect present invention embodiment, it is described according to the residual capacity
The prediction step predicted next time is adjusted, including:
Obtain the first ratio between the residual capacity and the total capacity of the disk;
By first ratio compared with default first threshold;
If first ratio exceeds the first threshold, the prediction step is determined according to the first numerical value;
If first ratio determines the prediction step without departing from the first threshold according to second value;Its
Described in the first numerical value be more than the second value.
A kind of possible implementation provided as first aspect present invention embodiment, methods described also include:
Since first time is trained to the forecast model of structure, the forecast model trained every time is judged
Each input weights whether be in respective confidential interval in;
If respectively input weights are in respective confidential interval, the forecast model that this is trained is carried out
Mark;
Count the number of the labeled forecast model;
Obtain the second ratio between the number of the labeled forecast model and currently accumulative total frequency of training;
If second ratio exceedes default Second Threshold, stop the training to the forecast model;
According to each input weights of all labeled forecast models, each input of the target prediction model is determined
Weights.
The acquisition methods of residual time length workable for the disk of the embodiment of the present invention, the monitoring of the multidimensional by gathering disk
Data are predicted model training as training data, after the completion of forecast model modeling, recycle the prediction data of multidimensional, defeated
Enter the prediction residual capacity to target prediction model to disk and carry out multi-step prediction, when the residual capacity predicted is zero, root
The prediction step of target prediction model determines the residual time length of disk when accordingly.Because prediction process uses the side of machine learning
Formula, and with reference to influence the multidimensional monitoring data of disk residual capacity, so as to improve the accuracy of prediction well.
For the above-mentioned purpose, the acquisition of residual time length workable for second aspect of the present invention embodiment proposes a kind of disk
Device, including:
Prediction module, for by the multidimensional prediction data input of disk into target prediction model, to obtain the disk
Residual capacity;
Adjusting module, for according to prediction step during residual capacity adjustment prediction next time;
Model modification module, for being updated according to the prediction step to the target prediction model;
The prediction module, for the prediction data to be re-entered into the target prediction model after renewal,
The residual capacity is predicted, and returns to the adjusting module and the prediction step, Yi Jisuo is adjusted according to the residual capacity
State model modification module to be updated the target prediction model according to the prediction step, the target after renewal
Untill the residual capacity that forecast model predicts is zero;
Determining module, the prediction step of the corresponding target prediction model is made during for being zero by the residual capacity
For residual time length workable for the disk.
A kind of possible implementation provided as second aspect of the present invention embodiment, the model modification module, tool
Body is used to obtain corresponding with the prediction step the first forecast model according to the prediction step, and by the target prediction
Model modification is first forecast model.
A kind of possible implementation provided as second aspect of the present invention embodiment, described device also include:
Training module, for according to the prediction step obtain corresponding with the prediction step the first forecast model it
Before, the multidimensional history monitoring data of disk is gathered as training data, in advance according to different prediction steps, utilizes the training
Data are trained to the forecast model of structure, obtain first forecast model corresponding with each prediction step.
A kind of possible implementation provided as second aspect of the present invention embodiment, the model modification module, tool
Body is used for using the multidimensional history monitoring data of the disk as training data, is re-entered into the forecast model of structure, presses
The forecast model is trained according to the prediction step, obtains the first forecast model corresponding with the prediction step, will
The target prediction model modification is first forecast model.
A kind of possible implementation provided as second aspect of the present invention embodiment, the adjusting module are specific to use
In obtaining the first ratio between the residual capacity and the total capacity of the disk, by first ratio and default first
Threshold value is compared, if first ratio exceeds the first threshold, the prediction step is determined according to the first numerical value,
And if first ratio then determines the prediction step without departing from the first threshold according to second value.
A kind of possible implementation provided as second aspect of the present invention embodiment, described device also include:
Judge module, for since first time is trained to the forecast model of structure, judgement to train every time
The forecast model each input weights whether be in respective confidential interval in;
Mark module, if be in for respectively inputting weights in respective confidential interval, the institute trained to this
Forecast model is stated to be marked;
Statistical module, for counting the number of the forecast model marked;
Acquisition module, for obtaining between the number of the forecast model marked and currently accumulative total frequency of training
The second ratio;
Stopping modular, if exceeding default Second Threshold for second ratio, stop to the forecast model
Training;
Weight determination module, for each input weights according to all labeled forecast models, determine the mesh
Mark each input weights of forecast model.
The acquisition device of residual time length workable for the disk of the embodiment of the present invention, the monitoring of the multidimensional by gathering disk
Data are predicted model training as training data, after the completion of forecast model modeling, recycle the prediction data of multidimensional, defeated
Enter the prediction residual capacity to target prediction model to disk and carry out multi-step prediction, when the residual capacity predicted is zero, root
The prediction step of target prediction model determines the residual time length of disk when accordingly.Because prediction process uses the side of machine learning
Formula, and with reference to influence the multidimensional monitoring data of disk residual capacity, so as to improve the accuracy of prediction well.
For the above-mentioned purpose, third aspect present invention embodiment proposes a kind of computer equipment, including:
Processor and memory;
Wherein, the processor by read the executable program code stored in the memory run with it is described can
Program corresponding to configuration processor code, for realizing residue workable for the disk as described in first aspect present invention embodiment
The acquisition methods of duration.
For the above-mentioned purpose, fourth aspect present invention embodiment proposes a kind of computer program product, when the calculating
When instruction in machine program product is by computing device, when performing remaining workable for the disk as described in first aspect embodiment
Long acquisition methods.
For the above-mentioned purpose, fifth aspect present invention embodiment proposes a kind of non-transitory computer-readable storage medium
Matter, computer program is stored thereon with, is realized when computer program is executed by processor as described in first aspect embodiment
The acquisition methods of residual time length workable for disk.
The additional aspect of the present invention and advantage will be set forth in part in the description, and will partly become from the following description
Obtain substantially, or recognized by the practice of the present invention.
Brief description of the drawings
Of the invention above-mentioned and/or additional aspect and advantage will become from the following description of the accompanying drawings of embodiments
Substantially and it is readily appreciated that, wherein:
Fig. 1 is that the flow of the acquisition methods of residual time length workable for a kind of disk provided in an embodiment of the present invention is illustrated
Figure;
Fig. 2 is a kind of schematic diagram of forecast model training process provided in an embodiment of the present invention;
Fig. 3 is a kind of structural representation of single hidden layer feedforward network provided in an embodiment of the present invention;
Fig. 4 is the application signal of the acquisition methods of residual time length workable for a kind of disk provided in an embodiment of the present invention
Figure;
Fig. 5 is the structural representation of the acquisition device of residual time length workable for a kind of disk provided in an embodiment of the present invention
Figure;
Fig. 6 is the structural representation of the acquisition device of residual time length workable for another disk provided in an embodiment of the present invention
Figure;
Fig. 7 is a kind of structural representation of computer equipment provided in an embodiment of the present invention.
Embodiment
Embodiments of the invention are described below in detail, the example of the embodiment is shown in the drawings, wherein from beginning to end
Same or similar label represents same or similar element or the element with same or like function.Below with reference to attached
The embodiment of figure description is exemplary, it is intended to for explaining the present invention, and is not considered as limiting the invention.
Below with reference to the accompanying drawings the acquisition methods and device of residual time length workable for describing the disk of the embodiment of the present invention.
Fig. 1 is the schematic flow sheet of the acquisition methods of residual time length workable for the disk of the embodiment of the present invention.Such as Fig. 1 institutes
Show, the acquisition of residual time length workable for the disk comprises the following steps:
S101, by the multidimensional prediction data input of disk into target prediction model, predict the remaining appearance of the disk
Amount.
, can be according to a machine learning algorithm one initial predicted model of structure in advance in the present embodiment, then can be with
The history monitoring data of disk is gathered, using history monitoring data as training data, the initial predicted model of structure is instructed
Practice, ultimately form the forecast model for being predicted to disk.
Wherein, machine learning algorithm can include:BP neural network, SVMs and single hidden layer feedforward network etc.,
Machine learning algorithm is not limited in the present embodiment.
Wherein, history monitoring data can be the monitoring data in preset time period, for example, the period can be one
The monitoring data of disk in month.Multidimensional history monitoring data as training data can include:Disk writing rate, per second write
Indegree, network interface card handling capacity, the multidimensional monitoring data such as average load of equipment where disk and each monitoring in preset duration
Disk residual capacity in cycle.Generally, it can be 15 minutes that preset duration, which is, and the monitoring cycle can be one day.
Herein it should be noted that on the training process to initial predicted model, it is introduced in subsequent embodiment, herein no longer
Repeat.
In the present embodiment, after final forecast model is got, it is possible to the residue of disk is held using forecast model
Amount measures.Specifically, the prediction data being predicted to disk can be gathered in advance, wherein, prediction data includes magnetic
The multidimensional such as average load of equipment where disk in disk writing rate, write-in number per second, network interface card handling capacity and preset duration
Monitoring data.Prediction data is input in current forecast model, prediction data carried out based on current forecast model pre-
Survey, obtain the residual capacity of disk.
S102, the prediction step predicted next time according to residual capacity adjustment.
In the present embodiment, it is used for for the prediction step of forecast model to embody the residual time length of disk, i.e. prediction step
Numerical value is the numerical value of residual time length.After residual capacity is got, first determine whether residual capacity is zero, if residual capacity
It is zero, illustrates that disk can continue to the residual time length corresponding to current forecast model, i.e., the prediction of current forecast model
Step-length.And if residual capacity non-zero, illustrate current forecast model prediction step can not when disk residual capacity be
Zero, that is to say, that after the residual time length that current forecast model predicts, disk still can be continuing with, and therefore, be needed
Again the residual capacity of disk is predicted, and the prediction step for needing adjustment to predict next time, that is, need toward up-regulation
Whole prediction step.For example, the prediction step of current forecast model is 5, i.e., the residual time length of disk is 5 days, when what is predicted
Residual capacity is not zero, and illustrates that disk at least can also be supported again by 1 day more.Therefore, it is necessary to up adjust what is predicted next time
Prediction step, the prediction step next time can be 6, or 8.
, can be according to the size of residual capacity, to determine to adjust the span of prediction step as a kind of example.Getting
After residual capacity, the numerical value for adjusting prediction step can be determined according to residual capacity and the total capacity of disk.Specifically,
Residual capacity and total capacity can be made ratio, obtain obtaining the first ratio between residual capacity and total capacity, then by first
Ratio is compared with default first threshold, if the first ratio exceeds first threshold, illustrates that the residual capacity of disk is larger,
In order to reduce the number of prediction, prediction step can be increased, improve prediction span, then prediction step can be determined according to the first numerical value
It is long.And if the first ratio without departing from first threshold, illustrates that the residual capacity of disk is smaller, in order to ensure the accuracy of prediction,
Need to predict by a small margin, then determine prediction step according to second value, wherein, the first numerical value is more than second value.Ordinary circumstance
Under, second value is arranged to 1.
S103, target prediction model is updated according to prediction step.
, can be different for different prediction step training previously according to the training data of collection as a kind of example
Forecast model, after prediction step is determined, it is possible to directly from training in advance come out multiple forecast models in, determine with
Target prediction model corresponding to the prediction step, is then updated to target prediction model by current forecast model.
As another example, by the training data of collection, it is re-entered into the initial predicted model of structure, then presses
Initial predicted model is trained according to prediction step, obtains target prediction model corresponding with prediction step, then will be current
Forecast model be updated to target prediction model.
In the present embodiment, whenever the residual capacity non-zero predicted, it is necessary to increase the prediction step of forecast model
It is long, so that the residual capacity that forecast model is exported is zero, and then obtain the residual time length of disk.
Herein it should be noted that training process on forecast model, is introduced, herein not in subsequent embodiment
Repeat again.
S104, multidimensional prediction data are re-entered into the target prediction model after renewal, predict residual capacity, and
Return to perform and prediction step is adjusted according to residual capacity, and target prediction model is updated according to prediction step, until
Untill the residual capacity that target prediction model prediction after renewal goes out is zero.
Can will be more after forecast model to be updated to target prediction model corresponding with prediction step in the present embodiment
Dimension prediction data is re-entered into forecast model, and the forecast model is the forecast model after updating.Based on pre- after renewal
Survey model can be predicted the residual capacity of disk, be then back to S102 and re-execute subsequent operation, until predicting mould
The residual capacity that type predicts is zero, when the residual capacity that forecast model predicts is zero, illustrates the residual time length of disk not
It can increase, therefore prediction step need not be increased, so as to stop the renewal to forecast model.
The prediction step of corresponding target prediction model remains as workable for disk when residual capacity is zero by S105
Remaining duration.
The acquisition methods of residual time length, the monitoring of the multidimensional by gathering disk workable for the disk that the present embodiment provides
Data are predicted model training as training data, after the completion of forecast model modeling, recycle the prediction data of multidimensional, defeated
Enter the prediction residual capacity to target prediction model to disk and carry out multi-step prediction, when the residual capacity predicted is zero, root
The prediction step of target prediction model determines the residual time length of disk when accordingly.Because prediction process uses the side of machine learning
Formula, and with reference to influence the multidimensional monitoring data of disk residual capacity, so as to improve the accuracy of prediction well.
The acquisition methods of residual time length, this reality workable for the disk provided for more clear explanation above-described embodiment
Apply in example and the training process of forecast model is introduced, as shown in Figure 2.Fig. 2 is a kind of prediction provided in an embodiment of the present invention
The schematic diagram of model training process.
The forecast model training process comprises the following steps:
S201, since first time is trained to the forecast model of structure, judge the forecast model trained every time
Whether each input weights are in respective confidential interval.
In the present embodiment, the model parameter of forecast model can be pre-set, and then form an initial predicted model.Mould
Shape parameter can include input weights, output weights, activation primitive, the number of plies and every node layer number etc..It is initial determining
After forecast model, the multidimensional history monitoring data in preset time period can be gathered as training data, wherein, preset time period
The multidimensional history monitoring data that interior each collection period collects, it is necessary to utilizes every group of instruction as one group of training data
Practice data to be trained to initial predicted model.For example, the multidimensional history monitoring number of daily disk in one month can be gathered
According to as training data.
As a kind of example, initial predicted model can be built using single hidden layer feedforward network.Single hidden layer feedforward network
Structural representation it is as shown in Figure 3.The expression formula of single hidden layer feedforward network is:
Wherein, N represents the number of training sample;ui=[ui1,ui2,…,uip]T∈RpRepresent that input vector p represents input
The dimension of vector;yj=[yj1,yj2,…,tjq]T∈RqOutput vector is represented, q represents the dimension of output vector;M is hidden node
Number;bjIt is implicit bigoted, g () is activation primitive;wj=[wj1,wj2,…,wjp]TTo input weights, p represents input vector
Dimension, βj=[βj1,βj2,…,βjq]TTo export weights, q represents the dimension of output weights.
It is each fixed respective confidential interval of input weights bidding in advance in the present embodiment, it is each in the training process to input power
Value is fallen into various confidential intervals, it may be said that bright forecast model meet demand.Forecast model is trained out from first time
Begin, whether each input weights for the first forecast model for judging to train every time are in respective confidential interval.If first
Each input weights of forecast model are in respective confidential interval, then perform S202, otherwise, the first forecast model are not entered
Line flag.
S202, the forecast model that this is trained is marked.
S203, count the number of labeled forecast model.
After carrying out the second forecast model mark to the first forecast model, the number of the second forecast model can be counted.Specifically
Ground, it can be counted using a counter, whenever one the second forecast model of mark, counter can be carried out adding 1.Will
Number of the count value of counter as the second forecast model.
S204, obtain the second ratio between the number of labeled forecast model and currently accumulative total frequency of training.
In the present embodiment, total frequency of training can be added up by another counter, then by the second forecast model
Number and total frequency of training make ratio, obtain the second ratio between the number of the second forecast model and total frequency of training.
S205, if the second ratio exceedes default Second Threshold, stop the training to forecast model.
When the second ratio exceeds default Second Threshold, the training to initial predicted model can be stopped.The present embodiment
In, Second Threshold can be 0.8, when the second ratio is 0.8, illustrate each defeated of the current forecast model having had beyond 80%
Enter weights to drop into confidential interval,, can be with by demarcating confidential interval so as to stop the training to initial predicted model
Improve the speed of training.
Herein it should be noted that after all training datas have been trained, if the second ratio is still not less than default
Second Threshold, then need to readjust confidential interval.
S206, according to each input weights of all labeled forecast models, it is determined that for entering to the residual capacity of disk
Each input weights of the forecast model of row prediction.
It is average it is possible to further which each input weights of all second forecast models are summed up, obtain each input
The average value of value, can be using the average value of each input weights as the prediction mould for being predicted to the residual capacity of disk
Each input weights of type.
In the present embodiment, input weights and corresponding confidential interval are set to each dimension monitoring data, in training process
In, the input weights of all dimension monitoring datas are all fallen into confidential interval beyond default threshold value, it is possible to deconditioning,
Training speed can be improved, and using the input weight average for falling into all second forecast models in confidential interval, is used
In the input weights of the forecast model of prediction, the accuracy that input weights determine is improved, so as to provide the accuracy of prediction.
Fig. 4 is that the application of the acquisition methods of residual time length workable for another disk provided in the embodiment of the present invention is shown
It is intended to.
S401, the multidimensional monitoring data in a period of time are obtained as training data.
Wherein, it can be for a period of time one month or some months.Multidimensional history monitoring data as training data can
With including:Disk writing rate, write-in number per second, network interface card handling capacity, in preset duration equipment where disk average load
Deng multidimensional monitoring data and the disk residual capacity in each monitoring cycle.
S402, the forecast model to be built based on single hidden layer feedforward network are set confidential interval, input weights and implied inclined
Hold.
Prediction, which can be based on, to be needed for respectively input weights set confidential interval, and can randomly set in forecast model
Input weights and imply bigoted.
S403, utilize training of the training data to forecast model.
Input weights based on setting and implicit bigoted, build initial forecast model, are walked when training first from prediction
A length of 0 starts to train, and being predicted the 1 step forecast model that step-length is 1 according to S403~S407 establishes.For example, predicted by 1 step
Model can complete the residual capacity based on data prediction today disk tomorrow;Differed between the input and output of training now
1 day.
S404, by each input weights of the forecast model trained every time compared with respective confidential interval.
S405, if each input weights of the forecast model this time trained are in respective confidential interval, to this
Secondary forecast model is marked.
S406, according to the number of labeled forecast model and currently accumulative total frequency of training, judges whether to meet institute
Having in input weights has 80% in confidential interval.
When judging to meet have 80% in confidential interval in all input weights, then deconditioning, performs S407, no
Then perform 421.
S407, generate target prediction model.
Specifically, each input weights of labeled forecast model are obtained, to each defeated of all labeled forecast models
Enter weights and sum up the average value for averagely, obtaining each input weights, formed using the average value of each input weights described
Target prediction model.
S408, prediction data is input in target prediction model, predicts the residual capacity of disk.
S409, judge whether residual capacity is zero.
If residual capacity is zero, S410 is performed;And if residual capacity is more than 0, illustrate to can use tomorrow, then residue makes
It with number of days+1, that is, now can at least use 1 day, then perform S411.
S410, by the remaining number of days that the prediction step of target prediction model is disk.
S411, prediction step is adjusted according to residual capacity, return and perform S402 to update the model knot of target prediction model
Structure, forecast model is trained again, with the target prediction model after being updated.
1 step forecast model can be obtained by being performed according to above-mentioned flow, and the residual capacity predicted due to 1 step forecast model is not
It is zero, then prediction step can be adjusted according to residual capacity.In the present embodiment, adjusted value can be set as 1, step will be predicted
Length is adjusted to 2, trains 2 step forecast models.That is, using data today, the day after tomorrow can be predicted based on 2 step forecast models
Residual capacity, now differed 2 days between the input and output of the training data of each input prediction model.
As a kind of example, if residual capacity is larger, adjusted value can be increased, can be with for example, it can be set to for 5
Multi-step Predictive Model is once trained by above-mentioned flow, i.e., can directly obtain 7 step forecast models from 2 step forecast models, from
And the time of prediction can be saved.
S412, judge when whether time training is most once to train.
If being trained for training for the last time when secondary, S402 is performed.Trained if not last time, then perform return
Perform S403.
In the present embodiment, input weights and corresponding confidential interval are set to each dimension monitoring data, in training process
In, the input weights of all dimension monitoring datas are all fallen into confidential interval beyond default threshold value, it is possible to deconditioning,
Training speed can be improved, and using the input weight average for falling into all second forecast models in confidential interval, is used
In the input weights of the forecast model of prediction, the accuracy that input weights determine is improved, so as to provide the accuracy of prediction.
Fig. 5 is the structural representation of the acquisition device of residual time length workable for a kind of disk provided in an embodiment of the present invention
Figure.As shown in figure 5, the acquisition device of residual time length includes workable for the disk:Prediction module 11, adjustment model 12, model
Update module 13 and determination model 14.
Prediction module 11, for by the multidimensional prediction data input of disk into target prediction model, to obtain the magnetic
The residual capacity of disk;
Adjusting module 12, for the prediction step predicted according to residual capacity adjustment next time;
Model modification module 13, for being updated according to the prediction step to the target prediction model;
Prediction module 11, for the prediction data to be re-entered into the target prediction model after renewal, in advance
The residual capacity is measured, and returns to the adjusting module and the prediction step is adjusted according to the residual capacity, and it is described
Model modification module is updated according to the prediction step to the target prediction model, and the target after renewal is pre-
Untill the residual capacity that survey model prediction goes out is zero;
Determining module 14, the prediction step of corresponding target prediction model during for being zero by the residual capacity
As residual time length workable for the disk.
Further, model modification module 13, specifically for being obtained and the prediction step pair according to the prediction step
The first forecast model answered, and by the target prediction model modification be first forecast model.
Further, model modification module 13, specifically for using the multidimensional history monitoring data of the disk as training
Data, be re-entered into the forecast model of structure, the forecast model be trained according to the prediction step, obtain with
First forecast model corresponding to the prediction step, it is first forecast model by the target prediction model modification.
Further, adjusting module 12, specifically for obtaining between the residual capacity and the total capacity of the disk
First ratio, by first ratio compared with default first threshold, if first ratio exceeds described first
Threshold value, then determine the prediction step according to the first numerical value, and if first ratio without departing from the first threshold, then
The prediction step is determined according to second value.
On Fig. 5 basis, Fig. 6 is obtaining for residual time length workable for another disk provided in an embodiment of the present invention
Take the structural representation of device.The acquisition device of residual time length also includes workable for the disk:Training module 15, judge module
16th, mark module 17, statistical module 18, acquisition module 19, stopping modular 20 and weight determination module 21.
Training module 15, for obtaining the first forecast model corresponding with the prediction step according to the prediction step
Before, the multidimensional history monitoring data of disk is gathered as training data, in advance according to different prediction steps, utilizes the instruction
Practice data to be trained the forecast model of structure, obtain first forecast model corresponding with each prediction step.
Judge module 16, for since first time is trained to the forecast model of structure, judging training every time
Whether each input weights of the forecast model gone out are in respective confidential interval.
Mark module 17, if be in for respectively inputting weights in respective confidential interval, this is trained
The forecast model is marked.
Statistical module 18, for counting the number of the forecast model marked.
Acquisition module 19, for obtain the number of the forecast model marked and currently accumulative total frequency of training it
Between the second ratio.
Stopping modular 20, if exceeding default Second Threshold for second ratio, stop to the prediction mould
The training of type.
Weight determination module 21, for each input weights according to all labeled forecast models, it is determined that described
Each input weights of target prediction model.
In the present embodiment, it is used as training data by gathering the monitoring data of multidimensional of disk and is predicted model training,
After the completion of forecast model modeling, the prediction data of multidimensional is recycled, it is remaining to be input to prediction of the target prediction model to disk
Capacity carries out multi-step prediction, when the residual capacity predicted is zero, is determined according to the prediction step of now target prediction model
Go out the residual time length of disk.Because prediction process is by the way of machine learning, and with reference to influence disk residual capacity
Multidimensional monitoring data, so as to improve the accuracy of prediction well.
Further, input weights and corresponding confidential interval are set to each dimension monitoring data, in the training process,
The input weights of all dimension monitoring datas are all fallen into confidential interval beyond default threshold value, it is possible to deconditioning, can
To improve training speed, and using the input weight average for falling into all second forecast models in confidential interval, it is used for
The input weights of the forecast model of prediction, the accuracy that input weights determine is improved, so as to provide the accuracy of prediction.
Fig. 7 shows the block diagram suitable for being used for the exemplary computer device 30 for realizing the application embodiment.Fig. 7 is shown
Computer equipment 30 be only an example, any restrictions should not be brought to the function and use range of the embodiment of the present application.
As shown in fig. 7, computer equipment 30 is showed in the form of universal computing device.The component of computer equipment 30 can be with
Including but not limited to:One or more processor or processing unit 31, system storage 32, connect different system component
The bus 33 of (including system storage 32 and processing unit 31).
Bus 33 represents the one or more in a few class bus structures, including memory bus or Memory Controller,
Peripheral bus, graphics acceleration port, processor or the local bus using any bus structures in a variety of bus structures.Lift
For example, these architectures include but is not limited to industry standard architecture (Industry Standard
Architecture;Hereinafter referred to as:ISA) bus, MCA (Micro Channel Architecture;Below
Referred to as:MAC) bus, enhanced isa bus, VESA (Video Electronics Standards
Association;Hereinafter referred to as:VESA) local bus and periphery component interconnection (Peripheral Component
Interconnection;Hereinafter referred to as:PCI) bus.
Computer equipment 30 typically comprises various computing systems computer-readable recording medium.These media can be it is any can be by
The usable medium that computer equipment 30 accesses, including volatibility and non-volatile media, moveable and immovable medium.
System storage 32 can include the computer system readable media of form of volatile memory, such as arbitrary access
Memory (Random Access Memory;Hereinafter referred to as:RAM) 40 and/or cache memory 41.Computer equipment 30
It may further include other removable/nonremovable, volatile/non-volatile computer system storage mediums.Only conduct
Citing, storage system 42 can be used for reading and writing immovable, non-volatile magnetic media, and (Fig. 7 do not show, commonly referred to as " hard disk
Driver ").Although not shown in Fig. 7, it can provide for the magnetic to may move non-volatile magnetic disk (such as " floppy disk ") read-write
Disk drive, and to removable anonvolatile optical disk (such as:Compact disc read-only memory (Compact Disc Read Only
Memory;Hereinafter referred to as:CD-ROM), digital multi read-only optical disc (Digital Video Disc Read Only
Memory;Hereinafter referred to as:DVD-ROM) or other optical mediums) read-write CD drive.In these cases, each driving
Device can be connected by one or more data media interfaces with bus 33.Memory 32 can include at least one program and produce
Product, the program product have one group of (for example, at least one) program module, and it is each that these program modules are configured to perform the application
The function of embodiment.
Program/utility 50 with one group of (at least one) program module 51, such as memory 32 can be stored in
In, such program module 51 includes --- but being not limited to --- operating system, one or more application program, other programs
Module and routine data, the realization of network environment may be included in each or certain combination in these examples.Program mould
Block 51 generally performs function and/or method in embodiments described herein.
Computer equipment 30 can also be with one or more external equipments 60 (such as keyboard, sensing equipment, display 70
Deng) communication, the equipment communication interacted with the computer equipment 30 can be also enabled a user to one or more, and/or with making
Obtain any equipment that the computer equipment 30 can be communicated with one or more of the other computing device (such as network interface card, modulatedemodulate
Adjust device etc.) communication.This communication can be carried out by input/output (I/O) interface 34.Also, computer equipment 30 may be used also
To pass through network adapter 35 and one or more network (such as LAN (Local Area Network;Hereinafter referred to as:
LAN), wide area network (Wide Area Network;Hereinafter referred to as:WAN) and/or public network, for example, internet) communication.Such as figure
Shown, network adapter 35 is communicated by bus 33 with other modules of computer equipment 30.It should be understood that although do not show in figure
Go out, computer equipment 30 can be combined and use other hardware and/or software module, included but is not limited to:Microcode, device drives
Device, redundant processing unit, external disk drive array, RAID system, tape drive and data backup storage system etc..
Processing unit 31 is stored in program in system storage 32 by operation, so as to perform various function application and
Data processing, such as the acquisition methods of residual time length workable for realizing the disk shown in Fig. 1-Fig. 2 and Fig. 4.
Any combination of one or more computer-readable media can be used.Computer-readable medium can be calculated
Machine readable signal medium or computer-readable recording medium.Computer-readable recording medium for example can be --- but it is unlimited
In system, device or the device of --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, or it is any more than combination.Calculate
The more specifically example (non exhaustive list) of machine readable storage medium storing program for executing includes:Electrical connection with one or more wires, just
Take formula computer disk, hard disk, random access memory (RAM), read-only storage (Read Only Memory;Hereinafter referred to as:
ROM), erasable programmable read only memory (Erasable Programmable Read Only Memory;Hereinafter referred to as:
EPROM) or flash memory, optical fiber, portable compact disc read-only storage (CD-ROM), light storage device, magnetic memory device or
Above-mentioned any appropriate combination.In this document, computer-readable recording medium can be any includes or storage program
Tangible medium, the program can be commanded the either device use or in connection of execution system, device.
Computer-readable signal media can include in a base band or as carrier wave a part propagation data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including --- but
It is not limited to --- electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be
Any computer-readable medium beyond computer-readable recording medium, the computer-readable medium can send, propagate or
Transmit for by instruction execution system, device either device use or program in connection.
The program code included on computer-readable medium can be transmitted with any appropriate medium, including --- but it is unlimited
In --- wireless, electric wire, optical cable, RF etc., or above-mentioned any appropriate combination.
Can with one or more programming languages or its combination come write for perform the application operation computer
Program code, described program design language include object oriented program language-such as Java, Smalltalk, C++,
Also include conventional procedural programming language-such as " C " language or similar programming language.Program code can be with
Fully perform, partly perform on the user computer on the user computer, the software kit independent as one performs, portion
Divide and partly perform or performed completely on remote computer or server on the remote computer on the user computer.
It is related in the situation of remote computer, remote computer can pass through the network of any kind --- including LAN (Local
Area Network;Hereinafter referred to as:) or wide area network (Wide Area Network LAN;Hereinafter referred to as:WAN) it is connected to user
Computer, or, it may be connected to outer computer (such as passing through Internet connection using ISP).
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or the spy for combining the embodiment or example description
Point is contained at least one embodiment or example of the present invention.In this manual, to the schematic representation of above-mentioned term not
Identical embodiment or example must be directed to.Moreover, specific features, structure, material or the feature of description can be with office
Combined in an appropriate manner in one or more embodiments or example.In addition, in the case of not conflicting, the skill of this area
Art personnel can be tied the different embodiments or example and the feature of different embodiments or example described in this specification
Close and combine.
In addition, term " first ", " second " are only used for describing purpose, and it is not intended that instruction or hint relative importance
Or the implicit quantity for indicating indicated technical characteristic.Thus, define " first ", the feature of " second " can be expressed or
Implicitly include at least one this feature.In the description of the invention, " multiple " are meant that at least two, such as two, three
It is individual etc., unless otherwise specifically defined.
Any process or method described otherwise above description in flow chart or herein is construed as, and represents to include
Module, fragment or the portion of the code of the executable instruction of one or more the step of being used to realize custom logic function or process
Point, and the scope of the preferred embodiment of the present invention includes other realization, wherein can not press shown or discuss suitable
Sequence, including according to involved function by it is basic simultaneously in the way of or in the opposite order, carry out perform function, this should be of the invention
Embodiment person of ordinary skill in the field understood.
Expression or logic and/or step described otherwise above herein in flow charts, for example, being considered use
In the order list for the executable instruction for realizing logic function, may be embodied in any computer-readable medium, for
Instruction execution system, device or equipment (such as computer based system including the system of processor or other can be held from instruction
The system of row system, device or equipment instruction fetch and execute instruction) use, or combine these instruction execution systems, device or set
It is standby and use.For the purpose of this specification, " computer-readable medium " can any can be included, store, communicate, propagate or pass
Defeated program is for instruction execution system, device or equipment or the dress used with reference to these instruction execution systems, device or equipment
Put.The more specifically example (non-exhaustive list) of computer-readable medium includes following:Electricity with one or more wiring
Connecting portion (electronic installation), portable computer diskette box (magnetic device), random access memory (RAM), read-only storage
(ROM), erasable edit read-only storage (EPROM or flash memory), fiber device, and portable optic disk is read-only deposits
Reservoir (CDROM).In addition, computer-readable medium, which can even is that, to print the paper of described program thereon or other are suitable
Medium, because can then enter edlin, interpretation or if necessary with it for example by carrying out optical scanner to paper or other media
His suitable method is handled electronically to obtain described program, is then stored in computer storage.
It should be appreciated that each several part of the present invention can be realized with hardware, software, firmware or combinations thereof.Above-mentioned
In embodiment, software that multiple steps or method can be performed in memory and by suitable instruction execution system with storage
Or firmware is realized.Such as, if realized with hardware with another embodiment, following skill well known in the art can be used
Any one of art or their combination are realized:With the logic gates for realizing logic function to data-signal from
Logic circuit is dissipated, the application specific integrated circuit with suitable combinational logic gate circuit, programmable gate array (PGA), scene can compile
Journey gate array (FPGA) etc..
Those skilled in the art are appreciated that to realize all or part of step that above-described embodiment method carries
Suddenly it is that by program the hardware of correlation can be instructed to complete, described program can be stored in a kind of computer-readable storage medium
In matter, the program upon execution, including one or a combination set of the step of embodiment of the method.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing module, can also
That unit is individually physically present, can also two or more units be integrated in a module.Above-mentioned integrated mould
Block can both be realized in the form of hardware, can also be realized in the form of software function module.The integrated module is such as
Fruit is realized in the form of software function module and as independent production marketing or in use, can also be stored in a computer
In read/write memory medium.
Storage medium mentioned above can be read-only storage, disk or CD etc..Although have been shown and retouch above
Embodiments of the invention are stated, it is to be understood that above-described embodiment is exemplary, it is impossible to be interpreted as the limit to the present invention
System, one of ordinary skill in the art can be changed to above-described embodiment, change, replace and become within the scope of the invention
Type.
Claims (10)
- A kind of 1. acquisition methods of residual time length workable for disk, it is characterised in that including:By the multidimensional prediction data input of disk into target prediction model, the residual capacity of the disk is predicted;Prediction step when being predicted according to residual capacity adjustment next time;Target prediction model is updated according to the prediction step;The multidimensional prediction data are re-entered into the target prediction model after renewal, predict the remaining appearance Amount, and return to execution and the prediction step is adjusted according to the residual capacity, and according to the prediction step to the target Forecast model is updated, untill the residual capacity that the target prediction model prediction after renewal goes out is zero;The prediction step of the corresponding target prediction model can be used as the disk when using the residual capacity being zero Residual time length.
- 2. according to the method for claim 1, it is characterised in that it is described according to the prediction step to the target prediction mould Type is updated, including:The first forecast model corresponding with the prediction step is obtained according to the prediction step;It is first forecast model by the target prediction model modification.
- 3. according to the method for claim 2, it is characterised in that described obtained according to the prediction step walks with the prediction Before target prediction model corresponding to length, in addition to:The multidimensional history monitoring data of disk is gathered as training data;In advance according to different prediction steps, the forecast model of structure is trained using the training data, obtain with often First forecast model corresponding to individual prediction step.
- 4. according to the method for claim 1, it is characterised in that it is described according to the prediction step to the target prediction mould Type is updated, including:Using the multidimensional history monitoring data of the disk as training data, it is re-entered into the forecast model of structure, according to The prediction step is trained to the forecast model, obtains the first forecast model corresponding with the prediction step;It is first forecast model by the target prediction model modification.
- 5. according to the method for claim 1, it is characterised in that described to adjust what is predicted next time according to the residual capacity Prediction step, including:Obtain the first ratio between the residual capacity and the total capacity of the disk;By first ratio compared with default first threshold;If first ratio exceeds the first threshold, the prediction step is determined according to the first numerical value;If first ratio determines the prediction step without departing from the first threshold according to second value;Wherein institute State the first numerical value and be more than the second value.
- 6. the method according to claim 3 or 4, it is characterised in that also include:Since first time is trained to the forecast model of structure, judge train every time the forecast model it is each Input whether weights are in respective confidential interval;If respectively input weights are in respective confidential interval, rower is entered to the forecast model that this is trained Note;Count the number of the labeled forecast model;Obtain the second ratio between the number of the labeled forecast model and currently accumulative total frequency of training;If second ratio exceedes default Second Threshold, stop the training to the forecast model;According to each input weights of all labeled forecast models, determine that each input of the target prediction model is weighed Value.
- A kind of 7. acquisition device of residual time length workable for disk, it is characterised in that including:Prediction module, for by the multidimensional prediction data input of disk into target prediction model, to obtain the surplus of the disk Covolume amount;Adjusting module, for according to prediction step during residual capacity adjustment prediction next time;Model modification module, for being updated according to the prediction step to the target prediction model;The prediction module, for the prediction data to be re-entered into the target prediction model after renewal, prediction Go out the residual capacity, and return to the adjusting module and the prediction step, and the mould are adjusted according to the residual capacity Type update module is updated according to the prediction step to the target prediction model, the target prediction after renewal Untill the residual capacity that model prediction goes out is zero;Determining module, the prediction step of the corresponding target prediction model is as institute during for being zero using the residual capacity Residual time length workable for stating disk.
- A kind of 8. computer equipment, it is characterised in that including:Processor and memory;Wherein, the processor can perform by reading the executable program code stored in the memory to run with described Program corresponding to program code, for realizing such as residual time length workable for the disk as described in any in claim 1-6 Acquisition methods.
- 9. a kind of computer program product, when the instruction in the computer program product is by computing device, perform as weighed The acquisition methods of residual time length workable for profit requires disk any one of 1-6.
- 10. a kind of non-transitorycomputer readable storage medium, is stored thereon with computer program, it is characterised in that the calculating The acquisition of residual time length workable for the disk as any one of claim 1-6 is realized when machine program is executed by processor Method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710601830.3A CN107480028B (en) | 2017-07-21 | 2017-07-21 | Method and device for acquiring usable residual time of disk |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710601830.3A CN107480028B (en) | 2017-07-21 | 2017-07-21 | Method and device for acquiring usable residual time of disk |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107480028A true CN107480028A (en) | 2017-12-15 |
CN107480028B CN107480028B (en) | 2020-09-18 |
Family
ID=60595167
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710601830.3A Active CN107480028B (en) | 2017-07-21 | 2017-07-21 | Method and device for acquiring usable residual time of disk |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107480028B (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108205424A (en) * | 2017-12-29 | 2018-06-26 | 北京奇虎科技有限公司 | Data migration method, device and electronic equipment based on disk |
CN108681496A (en) * | 2018-05-09 | 2018-10-19 | 北京奇艺世纪科技有限公司 | Prediction technique, device and the electronic equipment of disk failure |
CN109189323A (en) * | 2018-07-06 | 2019-01-11 | 华为技术有限公司 | Expansion method and equipment |
CN109376905A (en) * | 2018-09-20 | 2019-02-22 | 广东亿迅科技有限公司 | Disk space prediction technique, device, computer equipment and storage medium |
CN109885469A (en) * | 2019-02-27 | 2019-06-14 | 深信服科技股份有限公司 | A kind of expansion method, prediction model creation method, device, equipment and medium |
CN109976975A (en) * | 2019-03-26 | 2019-07-05 | 北京大道云行科技有限公司 | A kind of disk size prediction technique, device, electronic equipment and storage medium |
CN110377228A (en) * | 2019-06-19 | 2019-10-25 | 深圳壹账通智能科技有限公司 | Automatic expansion method, device, O&M terminal and the storage medium of block chain node |
CN111752903A (en) * | 2020-06-23 | 2020-10-09 | 深圳前海微众银行股份有限公司 | Method for predicting available time of data storage space |
CN113687777A (en) * | 2021-07-23 | 2021-11-23 | 苏州浪潮智能科技有限公司 | Method, device, equipment and medium for predicting usable time of disk |
CN113835626A (en) * | 2021-09-02 | 2021-12-24 | 深圳前海微众银行股份有限公司 | Method and device for determining usable time of disk |
CN114327241A (en) * | 2020-09-29 | 2022-04-12 | 伊姆西Ip控股有限责任公司 | Method, electronic device and computer program product for managing disk |
CN114594909A (en) * | 2022-03-02 | 2022-06-07 | 北京金山云网络技术有限公司 | Disk processing method and device, electronic equipment and storage medium |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030225563A1 (en) * | 2002-05-30 | 2003-12-04 | Gonos Dan G. | Capacity planning |
CN102968573A (en) * | 2012-12-14 | 2013-03-13 | 哈尔滨工业大学 | Online lithium ion battery residual life predicting method based on relevance vector regression |
CN103033761A (en) * | 2012-12-17 | 2013-04-10 | 哈尔滨工业大学 | Lithium ion battery residual life forecasting method of dynamic gray related vector machine |
CN103942457A (en) * | 2014-05-09 | 2014-07-23 | 浙江师范大学 | Water quality parameter time series prediction method based on relevance vector machine regression |
US20140330793A1 (en) * | 2011-06-30 | 2014-11-06 | Emc Corporation | Capacity forecasting for a deduplicating storage system |
CN104899658A (en) * | 2015-06-12 | 2015-09-09 | 哈尔滨工业大学 | Prediction model selection method based on applicability quantification of time series prediction model |
CN106201849A (en) * | 2016-07-01 | 2016-12-07 | 南京航空航天大学 | Longevity Forecasting Methodology more than the long-life parts that a kind of finite data drives |
US9646256B2 (en) * | 2012-06-29 | 2017-05-09 | EMC IP Holding Company LLC | Automated end-to-end sales process of storage appliances of storage systems using predictive modeling |
-
2017
- 2017-07-21 CN CN201710601830.3A patent/CN107480028B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030225563A1 (en) * | 2002-05-30 | 2003-12-04 | Gonos Dan G. | Capacity planning |
US20140330793A1 (en) * | 2011-06-30 | 2014-11-06 | Emc Corporation | Capacity forecasting for a deduplicating storage system |
US9646256B2 (en) * | 2012-06-29 | 2017-05-09 | EMC IP Holding Company LLC | Automated end-to-end sales process of storage appliances of storage systems using predictive modeling |
CN102968573A (en) * | 2012-12-14 | 2013-03-13 | 哈尔滨工业大学 | Online lithium ion battery residual life predicting method based on relevance vector regression |
CN103033761A (en) * | 2012-12-17 | 2013-04-10 | 哈尔滨工业大学 | Lithium ion battery residual life forecasting method of dynamic gray related vector machine |
CN103942457A (en) * | 2014-05-09 | 2014-07-23 | 浙江师范大学 | Water quality parameter time series prediction method based on relevance vector machine regression |
CN104899658A (en) * | 2015-06-12 | 2015-09-09 | 哈尔滨工业大学 | Prediction model selection method based on applicability quantification of time series prediction model |
CN106201849A (en) * | 2016-07-01 | 2016-12-07 | 南京航空航天大学 | Longevity Forecasting Methodology more than the long-life parts that a kind of finite data drives |
Non-Patent Citations (2)
Title |
---|
史华洁: "《基于灰色神经网络的设备剩余寿命在线预测》", 《中国优秀硕士学位论文全文数据库》 * |
尤明懿: "《基于相似性的剩余寿命预测:鲁棒性与不确定性研究》", 《电子产品可靠性与环境试验》 * |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108205424A (en) * | 2017-12-29 | 2018-06-26 | 北京奇虎科技有限公司 | Data migration method, device and electronic equipment based on disk |
CN108681496A (en) * | 2018-05-09 | 2018-10-19 | 北京奇艺世纪科技有限公司 | Prediction technique, device and the electronic equipment of disk failure |
CN109189323A (en) * | 2018-07-06 | 2019-01-11 | 华为技术有限公司 | Expansion method and equipment |
CN109376905A (en) * | 2018-09-20 | 2019-02-22 | 广东亿迅科技有限公司 | Disk space prediction technique, device, computer equipment and storage medium |
CN109885469B (en) * | 2019-02-27 | 2022-09-30 | 深信服科技股份有限公司 | Capacity expansion method, prediction model creation method, device, equipment and medium |
CN109885469A (en) * | 2019-02-27 | 2019-06-14 | 深信服科技股份有限公司 | A kind of expansion method, prediction model creation method, device, equipment and medium |
CN109976975A (en) * | 2019-03-26 | 2019-07-05 | 北京大道云行科技有限公司 | A kind of disk size prediction technique, device, electronic equipment and storage medium |
CN109976975B (en) * | 2019-03-26 | 2023-07-28 | 北京大道云行科技有限公司 | Disk capacity prediction method and device, electronic equipment and storage medium |
CN110377228A (en) * | 2019-06-19 | 2019-10-25 | 深圳壹账通智能科技有限公司 | Automatic expansion method, device, O&M terminal and the storage medium of block chain node |
CN111752903A (en) * | 2020-06-23 | 2020-10-09 | 深圳前海微众银行股份有限公司 | Method for predicting available time of data storage space |
CN111752903B (en) * | 2020-06-23 | 2024-06-07 | 深圳前海微众银行股份有限公司 | Prediction method for usable time of data storage space |
CN114327241A (en) * | 2020-09-29 | 2022-04-12 | 伊姆西Ip控股有限责任公司 | Method, electronic device and computer program product for managing disk |
CN113687777A (en) * | 2021-07-23 | 2021-11-23 | 苏州浪潮智能科技有限公司 | Method, device, equipment and medium for predicting usable time of disk |
CN113687777B (en) * | 2021-07-23 | 2023-08-08 | 苏州浪潮智能科技有限公司 | Method, device, equipment and medium for predicting usable time length of magnetic disk |
WO2023029680A1 (en) * | 2021-09-02 | 2023-03-09 | 深圳前海微众银行股份有限公司 | Method and apparatus for determining usable duration of magnetic disk |
CN113835626B (en) * | 2021-09-02 | 2024-04-05 | 深圳前海微众银行股份有限公司 | Method and device for determining usable time length of disk |
CN113835626A (en) * | 2021-09-02 | 2021-12-24 | 深圳前海微众银行股份有限公司 | Method and device for determining usable time of disk |
CN114594909A (en) * | 2022-03-02 | 2022-06-07 | 北京金山云网络技术有限公司 | Disk processing method and device, electronic equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN107480028B (en) | 2020-09-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107480028A (en) | The acquisition methods and device of residual time length workable for disk | |
CN109543909A (en) | Prediction technique, device and the computer equipment of vehicle caseload | |
CN107133699A (en) | Box office receipts trend forecasting method and device, equipment, storage medium | |
CN110794308B (en) | Method and device for predicting train battery capacity | |
CN107464023A (en) | The on-line study method and apparatus of life cycle forecast model | |
CN104346926B (en) | Running time Forecasting Methodology and device and related terminal device | |
CN104361415B (en) | A kind of choosing method and device for showing information | |
CN111091196A (en) | Passenger flow data determination method and device, computer equipment and storage medium | |
CN108182490A (en) | A kind of short-term load forecasting method under big data environment | |
CN103399280A (en) | Method for predicting cycle life of lithium ion battery based on NSDP-AR (AutoRegressive) model | |
CN113821903B (en) | Temperature control method and equipment, modularized data center and storage medium | |
CN107357764A (en) | Data analysing method, electronic equipment and computer-readable storage medium | |
CN110264270A (en) | A kind of behavior prediction method, apparatus, equipment and storage medium | |
CN107274215A (en) | Flight prices Forecasting Methodology, device, equipment and storage medium | |
CN107729473A (en) | Article recommends method and its device | |
CN113222403A (en) | Power adjusting method and device based on big data, storage medium and electronic equipment | |
CN115994611A (en) | Training method, prediction method, device and storage medium for category prediction model | |
CN116842811B (en) | Workshop production line energy consumption analysis method and device based on digital twin and storage medium | |
CN113554184A (en) | Model training method and device, electronic equipment and storage medium | |
CN113033903A (en) | Fruit price prediction method, medium and equipment of LSTM model and seq2seq model | |
CN103279549B (en) | A kind of acquisition methods of target data of destination object and device | |
CN108090625A (en) | A kind of Forecasting Methodology, terminal device and the storage medium of finance market | |
CN107886180A (en) | Boat department creates single interface monitoring method, apparatus, electronic equipment, storage medium | |
CN103763123A (en) | Method and device for evaluating health condition of network | |
CN107169585A (en) | Film box office Forecasting Methodology, device and storage medium based on artificial intelligence |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |