CN108075906A - A kind of management method and system for cloud computation data center - Google Patents
A kind of management method and system for cloud computation data center Download PDFInfo
- Publication number
- CN108075906A CN108075906A CN201610980467.6A CN201610980467A CN108075906A CN 108075906 A CN108075906 A CN 108075906A CN 201610980467 A CN201610980467 A CN 201610980467A CN 108075906 A CN108075906 A CN 108075906A
- Authority
- CN
- China
- Prior art keywords
- index
- managed
- achievement data
- abnormality detection
- abnormality
- 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
Links
- 238000007726 management method Methods 0.000 title claims abstract description 30
- 230000005856 abnormality Effects 0.000 claims abstract description 119
- 238000001514 detection method Methods 0.000 claims abstract description 92
- 238000012549 training Methods 0.000 claims abstract description 48
- 238000000034 method Methods 0.000 claims abstract description 17
- 230000002159 abnormal effect Effects 0.000 claims description 11
- 238000012544 monitoring process Methods 0.000 abstract description 15
- 238000012423 maintenance Methods 0.000 abstract description 12
- 230000003068 static effect Effects 0.000 abstract description 9
- 238000012545 processing Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000013450 outlier detection Methods 0.000 description 1
- 230000002688 persistence Effects 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/04—Network management architectures or arrangements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0631—Management of faults, events, alarms or notifications using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Environmental & Geological Engineering (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of management methods and system for cloud computation data center.This method includes：The index to be managed in the cloud computation data center that user determines is obtained, and obtains the achievement data of the index to be managed；Whether the abnormality detection model based on structure determines the achievement data in abnormality；If the achievement data is in abnormality, the warning information of the index to be managed is exported with setting means.Utilize this method, it can be based on the abnormality detection model that training study obtains, the autonomous abnormality for treating level of control is accurately detected, and quickly carry out pre-alert notification, the troublesome operation that operation maintenance personnel is needed to be each index allocation static threshold to be managed in traditional monitoring warning system is avoided, so as to reduce the operation management resource input that data center is calculated cloud formula.
Description
Technical field
The present embodiments relate to field of cloud computer technology more particularly to a kind of managers for cloud computation data center
Method and system.
Background technology
Cloud computing is a kind of calculation based on internet, in this way, shared software and hardware resources and information
Computer and other equipment can be supplied on demand.Briefly, cloud computation data center is equivalent to by substantial amounts of physical machine (clothes
Be engaged in device) or the computing devices such as virtual machine composition support cloud computing computing device resource.With the development of internet, Hen Duogong
The portfolio of especially Internet company of department increases severely, and in order to support business Effec-tive Function, own data have been built by these companies
Center realizes the cloud deployment of business.
In order to ensure the reliable and stable operation of service application, many enterprises can all build a set of monitoring warning system to realize
Management to cloud computation data center.In traditional monitoring warning system, finger to be monitored in cloud computation data center is being determined
, it is necessary to which corresponding operation maintenance personnel manual setting sets the static alarm threshold of each resource metrics after mark, cycle detection is then carried out,
Alarm operation is carried out to the index for reaching threshold value.
It is above-mentioned to be had the following disadvantages based on static alarm threshold method：1st, configuration is cumbersome, and operation maintenance personnel is needed in each prison
The configuration of static alarm threshold is carried out on control host to definite index respectively, due to possessed by cloud computation data center
Server, virtual machine quantity are very big, and manual configuration cost is very high；2nd, need operation maintenance personnel have comparable system O＆M,
Business O＆M professional knowledge carries out rational static alarm threshold setting, otherwise, threshold value easily occurs and set excessively high leakage to be caused to be accused
The setting of alert or threshold value is too low to be caused to alert by mistake, and only after failure generation, operation maintenance personnel just can know that, cause in business
Disconnected operation.
The content of the invention
The present invention provides a kind of management method and system for cloud computation data center, to realize in terms of to existing cloud
The optimization to count according to center monitoring warning system.
The embodiment of the present invention uses following technical scheme：
In a first aspect, an embodiment of the present invention provides a kind of management method for cloud computation data center, this method bags
It includes：
The index to be managed in the cloud computation data center that user determines is obtained, and obtains the index of the index to be managed
Data；
Whether the abnormality detection model based on structure determines the achievement data in abnormality；
If the achievement data is in abnormality, the alarm that the index to be managed is exported with setting means is believed
Breath.
Second aspect, the embodiment of the present invention additionally provide a kind of management system for cloud computation data center, the system
Including：
Achievement data acquisition module for obtaining the index to be managed in the cloud computation data center that user determines, and obtains
Take the achievement data of the index to be managed；
Abnormal state detection module for the abnormality detection model based on structure, determines whether the achievement data is in
Abnormality；
Warning information output module, for when the achievement data is in abnormality, exported with setting means described in
The warning information of index to be managed.
The present invention provides a kind of management method and system for cloud computation data center, which determines first
Index to be managed in cloud computation data center, and obtain the achievement data of the index to be managed；It is then based on the different of structure
Whether normal detection model determines the achievement data in abnormality；Finally when the achievement data is in abnormality,
The warning information of the index to be managed is exported with setting means.The technical solution carried using the present invention can be based on training
Learn obtained abnormality detection model, the autonomous abnormality for treating level of control is accurately detected, and is quickly carried out pre-
Alert notice, avoiding needs operation maintenance personnel in traditional monitoring warning system be the cumbersome of each index allocation static threshold to be managed
Operation, so as to reduce the operation management resource input for calculating cloud formula data center.
Description of the drawings
Fig. 1 is a kind of flow chart for management method for cloud computation data center that the embodiment of the present invention one provides；
Fig. 2 is a kind of flow chart of management method for cloud computation data center provided by Embodiment 2 of the present invention；
Fig. 3 is a kind of structure chart for managing device for cloud computation data center that the embodiment of the present invention three provides.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention rather than limitation of the invention.Exemplary reality is being discussed in greater detail
It should be mentioned that some exemplary embodiments are described as the processing described as flow chart or method before applying example.Although
Operations (or step) are described as the processing of order by flow chart, but many of which operation can be by concurrently, concurrently
Ground is implemented simultaneously.In addition, the order of operations can be rearranged.The processing when its operations are completed can be by
It terminates, it is also possible to have the additional step being not included in attached drawing.It is described processing can correspond to method, function, regulation,
Subroutine, subprogram etc..
Embodiment one
Fig. 1 is a kind of flow chart for management method for cloud computation data center that the embodiment of the present invention one provides, this
Embodiment is applicable to be monitored indices in cloud computation data center the situation of alarm management, this method can by with
It is performed in the management system of cloud computation data center.The system can be realized by way of hardware and/or software, and generally may be used
It is integrated in the service platform for carrying out cloud computing.
As shown in Figure 1, a kind of management method for cloud computation data center that the embodiment of the present invention one provides, specific to wrap
It includes：
Index to be managed in the cloud computation data center that S101, acquisition user determine, and obtain the index to be managed
Achievement data.
In the present embodiment, the cloud computation data center can specifically regard as by substantial amounts of service host or virtual robot arm into
Computing device resource center, the index to be managed specifically can refer in cloud computation data center the hardware of user's expectation management or
Software index can be each service host to be managed or service index (such as service host of virtual machine in cloud computation data center
CUP occupancies, the Current Temperatures of CPU or memory usage etc.)；It can also be user configuration in cloud computation data center
On related application service index.Usually, treated described in being determined according to operation of the user in monitoring interface is set
Level of control.
It in the present embodiment, can be by being based on setting data acquisition side after acquisition determines the index to be managed
Formula monitors and obtains the achievement data of the index to be managed.Illustratively, the monitoring of resource occupation monitoring program can be based on simultaneously
The corresponding CPU of service host to be managed of acquisition or the occupancy of memory, are also based on the temperature sensor monitors set and adopt
Collect the Current Temperatures information of cloud computation data center.Usually, the present embodiment can gather the index to be managed in real time
Achievement data can also carry out the acquisition of achievement data based on certain time interval.Further, since in cloud computation data center
With substantial amounts of computing device resource, it is possible thereby to determine that the quantity of the index to be managed is also relatively large.
Whether S102, the abnormality detection model based on structure determine the achievement data in abnormality.
In the present embodiment, can abnormality be carried out to acquired achievement data by the abnormality detection model of structure
Determine.The abnormality detection model can specifically be trained by abnormality detection machine learning algorithm by ceaselessly algorithm iteration
Learn and build and formed, the abnormality detection model built after training study can form to be managed in the cloud computation data center
The Standard Eigenvalue of index.
In the present embodiment, after the abnormality detection model is started, achievement data can be inputted with data-stream form
The abnormality detection model, then can be by the Standard Eigenvalue phase of the characteristic value that the achievement data generates and index to be managed
It compares, to determine whether the achievement data is in abnormality.Specifically, if the achievement data generate characteristic value not
Characteristic value is complied with standard, then it is believed that the achievement data is in abnormality.
It should be noted that since the achievement data is inputted with data-stream form, and have in cloud computation data center
Substantial amounts of index to be managed, is thus taken into account and achievement data is carried out abnormality detection based on distributed streaming computing platform, example
Property, common distributive type computing platform has the streaming computings platform such as SparkStreaming, Storm.
If S103, the achievement data are in abnormality, the announcement of the index to be managed is exported with setting means
Alert information.
In the present embodiment, when detecting that the achievement data is in abnormality based on step S102, it may be determined that
The current working status of index to be managed is abnormal, it is necessary to carry out early warning processing in the cloud computation data center, specifically, can be with
The mode of mail or short message answers the warning information of index to be managed to operation maintenance personnel the output phase.
A kind of management method for cloud computation data center that the embodiment of the present invention one provides, it is first determined cloud computing number
According to the index to be managed in center, and obtain the achievement data of the index to be managed；It is then based on the abnormality detection mould of setting
Whether type determines the achievement data in abnormality；Finally when the achievement data is in abnormality, with the side of setting
Formula exports the warning information of the index to be managed.The technical solution carried using the present invention can be based on training study and obtain
Abnormality detection model, the autonomous abnormality for treating level of control is accurately detected, and quickly carries out pre-alert notification, is kept away
Exempt from the troublesome operation that operation maintenance personnel is needed to be each index allocation static threshold to be managed in traditional monitoring warning system, so as to
Reduce the operation management resource input that data center is calculated cloud formula.
Embodiment two
Fig. 2 is a kind of flow chart of management method for cloud computation data center provided by Embodiment 2 of the present invention.This
Inventive embodiments two are optimized based on above-described embodiment, in the present embodiment, will obtain the finger of the index to be managed
Data are marked, are specifically optimized for：Acquisition or the achievement data with index to be managed described in setting time interval acquiring in real time；Accordingly
, the present embodiment, which also optimizes, to be added：The achievement data of the index to be managed is stored in the database of setting, forms institute
State the history achievement data of index to be managed.
Further, whether the abnormality detection model based on structure is determined the achievement data in different by the present embodiment
Normal state, is specifically optimized for：Using the achievement data as the input value of the abnormality detection model of structure, corresponding index is generated
Data characteristics；If the achievement data feature meets the standard index feature of the abnormality detection model, it is determined that the finger
Mark data are in normal condition；Otherwise, it determines the achievement data is in abnormality.
Further, the present embodiment also optimizes and adds：Learning algorithm, distributed structure are trained according to the abnormality detection of setting
Build the abnormality detection model of the cloud computation data center.
As shown in Fig. 2, a kind of management method for cloud computation data center provided by Embodiment 2 of the present invention, specific to wrap
Include following operation：
Index to be managed in the cloud computation data center that S201, acquisition user determine.
Illustratively, the index option that operation maintenance personnel is chosen in the monitoring interface of monitoring host computer can be obtained, it will be above-mentioned
Index option is determined as the index to be managed in cloud computation data center.For example, the index to be managed can be cloud computing number
According to the hardware of service host or virtual machine in running order in center or software index；It can also be user configuration in cloud meter
The service index to count according to supercentral application program.
S202, in real time acquisition or the achievement data with setting time interval acquiring index to be managed, and achievement data is deposited
Storage forms the history achievement data of index to be managed in the database of setting.
In the present embodiment, acquired achievement data is often present in the instant data in memory, the achievement data
The monitoring management of abnormality is can be directly used for, persistence operation can also be carried out to achievement data, achievement data is deposited in
In specified data storehouse, permanent history achievement data is formed, which can be used for follow-up abnormality detection model
Training study.It is understood that the present embodiment can monitor and obtain the achievement data of index to be managed in real time, thus
Embody the real-time that the present embodiment provides management method；In addition, when the data variation frequency of the index to be managed is relatively low,
It is contemplated that the acquisition achievement data at interval, thus on the premise of this implementation management method accuracy is not influenced, to reduce fortune
Resource consumption during row.
S203, learning algorithm is trained according to the abnormality detection of setting, distribution builds the different of the cloud computation data center
Normal detection model.
It should be noted that step S203 is equivalent to the structure operation of an abnormality detection model, with step S201 and step
Rapid 202 realization is limited without priority, as long as being realized before step S204 is performed.
In the present embodiment, since the quantity of identified index to be managed in cloud computation data center is larger, thus examine
Consider and build distributed abnormality detection model, to support the Outlier Detection Algorithm iteration of big data quantity, while accelerate to be managed
The monitoring of Indexes Abnormality state.In the present embodiment, abnormality detection training learning algorithm can be but be not limited to one kind
The existing abnormality detection learning algorithm such as support vector machines (One-Class SVM) and Multi-dimensional Gaussian distribution.
Further, described to train learning algorithm according to the abnormality detection of setting, distribution builds the cloud computing data
The abnormality detection model at center includes：The abnormality detection training learning algorithm of setting is disposed on Distributed Computing Platform；From number
According to the history achievement data that the index to be managed is obtained in storehouse, the training data as abnormality detection training learning algorithm
Collection；Determine the index feature of the index to be managed, the training learning characteristic as abnormality detection training learning algorithm；Base
In the training dataset and the trained learning characteristic, train and obtained in the cloud computing data on Distributed Computing Platform
The abnormality detection model of the heart.
In the present embodiment, in the building process of abnormality detection model, it is necessary first to be disposed in Distributed Computing Platform
The abnormality detection training learning algorithm of setting, wherein, common Distributed Computing Platform has Spark and Hadoop platform etc.；So
Going through for the index to be managed can be obtained in the slave database based on Scheduling Framework Oozie or scheduling system Cron timings afterwards
History achievement data, using the training dataset as abnormality detection training learning algorithm, wherein, the history obtained from database refers to
Mark data can be the data of one month, three months or longer time, and specific duration does not limit, as long as training can be embodied
The practicability of training data in data set；It also needs to determine the training learning characteristic needed for training study afterwards, usually,
The training learning characteristic can come from the index feature of index to be managed, could also be from the training characteristics of user's sets itself；
Training dataset and training learning characteristic may finally be based on, the training acquisition cloud computing data on Distributed Computing Platform
The abnormality detection model at center.
S204, using the achievement data of index to be managed as structure abnormality detection model input value, generate it is corresponding
Achievement data feature.
In the present embodiment, the judgement of the abnormality of achievement data has been described in detail in step S204 and step S205
Journey.Specifically, the achievement data of index to be managed can be inputted based on distributive type computing platform to abnormality detection model,
Corresponding achievement data feature can be generated based on the abnormality detection model, can be sentenced afterwards based on the achievement data feature
Determine the abnormality of achievement data.
If S205, achievement data feature meet the standard index feature of abnormality detection model, it is determined that the index number
According in normal condition；Otherwise, it determines the achievement data is in abnormality.
In the present embodiment, the standard index feature specifically can be regarded as what the operation training based on step S203 obtained
The normal index feature that index in cloud computation data center should meet.Therefore, when the achievement data character symbol of achievement data
When closing the standard index feature of abnormality detection model, it is believed that achievement data is in normal condition.It is understood that based on upper
State operation, can be independently learnt based on training after abnormality detection model quickly and accurately realize and treat the exception of level of control
Monitoring.
If S206, the achievement data are in abnormality, the announcement of the index to be managed is exported with setting means
Alert information.
Illustratively, when the achievement data monitored is in abnormality, it is possible to which the short message mode of setting is to O＆M
Personnel's the output phase answers the warning information of index to be managed, so that operation maintenance personnel can timely carry out abnormal index to be managed
It safeguards.
A kind of management method for cloud computation data center provided by Embodiment 2 of the present invention, embodies abnormality detection
The construction process of model, while embody the decision process of abnormality.Using this method, abnormality detection machine can be based on
The thought of study is gone out the normal model of cloud computation data center by ceaselessly algorithm iteration automatic modeling, avoids traditional prison
Control the troublesome operation of a large amount of static alarm threshold of operation maintenance personnel manual setting in warning system；In addition, based on Distributed Calculation
Thought, to each New Set data by the judgement of distributed abnormality detection model, precise and high efficiency realizes abnormality
Monitoring and alarm, timely avoid the influence that service disconnection is brought, so as to reduce the O＆M for calculating cloud formula data center
Manage resource input.
Embodiment three
Fig. 3 is a kind of structure chart for managing device for cloud computation data center that the embodiment of the present invention three provides.It should
System is applicable to be monitored indices in cloud computation data center the situation of alarm management, can by hardware and/or
The mode of software is realized, and can be generally integrated in the service platform for carrying out cloud computing.As shown in figure 3, the system includes：Index
Data acquisition module 31, abnormal state detection module 32 and warning information output module 33.
Wherein, achievement data acquisition module 31, for obtaining the finger to be managed in the cloud computation data center that user determines
Mark, and obtain the achievement data of the index to be managed；
Abnormal state detection module 32 for the abnormality detection model based on structure, determines whether the achievement data is located
In abnormality；
Warning information output module 33, for when the achievement data is in abnormality, institute to be exported with setting means
State the warning information of index to be managed.
In the present embodiment, which determines to treat in cloud computation data center by achievement data acquisition module 31 first
Level of control, and obtain the achievement data of the index to be managed；Then by abnormal state detection module 32 based on setting
Whether abnormality detection model determines the achievement data in abnormality；Work as institute eventually by warning information output module 33
When stating achievement data and being in abnormality, the warning information of the index to be managed is exported with setting means.
A kind of management system for cloud computation data center that the embodiment of the present invention three provides, passes through achievement data first
Acquisition module obtains the achievement data of index to be managed；Then by abnormal state detection module monitors achievement data it is different
Normal state；The alarm prompt of abnormality has been carried out eventually by warning information output module.Using the system, instruction can be based on
Practice the abnormality detection model that study obtains, the autonomous abnormality for treating level of control is accurately detected, and is quickly carried out
Pre-alert notification, avoiding needs operation maintenance personnel in traditional monitoring warning system be the numerous of each index allocation static threshold to be managed
Trivial operation, so as to reduce the operation management resource input for calculating cloud formula data center.
Further, which, which also optimizes, includes：
Detection model build module 34, for according to the abnormality detection of setting train learning algorithm, distribution structure described in
The abnormality detection model of cloud computation data center.
On the basis of above-described embodiment, the detection model structure module 34 is specifically used for：
The abnormality detection training learning algorithm of setting is disposed on Distributed Computing Platform；It is treated described in being obtained from database
The history achievement data of level of control, the training dataset as abnormality detection training learning algorithm；Pipe is treated described in determining
The index feature of index is managed, the training learning characteristic as abnormality detection training learning algorithm；Based on the training data
Collection and the trained learning characteristic, the abnormality detection mould of the training acquisition cloud computation data center on Distributed Computing Platform
Type.
Further, obtaining the achievement data of the index to be managed includes：
Acquisition or the achievement data with index to be managed described in setting time interval acquiring in real time；
Correspondingly, the system also includes：
History data store module 35, for the achievement data of the index to be managed to be stored in the database of setting
In, the history achievement data of the formation index to be managed.
Further, abnormal state detection module 32 is specifically used for：
Using the achievement data as the input value of the abnormality detection model of structure, corresponding achievement data feature is generated；
If the achievement data feature meets the standard index feature of the abnormality detection model, it is determined that the achievement data is in
Normal condition；Otherwise, it determines the achievement data is in abnormality.
Note that it above are only presently preferred embodiments of the present invention and institute's application technology principle.It will be appreciated by those skilled in the art that
The invention is not restricted to specific embodiment described here, can carry out for a person skilled in the art various apparent variations,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out by above example to the present invention
It is described in further detail, but the present invention is not limited only to above example, without departing from the inventive concept, also
It can include other more equivalent embodiments, and the scope of the present invention is determined by scope of the appended claims.
Claims (10)
1. a kind of management method for cloud computation data center, which is characterized in that including：
The index to be managed in the cloud computation data center that user determines is obtained, and obtains the index number of the index to be managed
According to；
Whether the abnormality detection model based on structure determines the achievement data in abnormality；
If the achievement data is in abnormality, the warning information of the index to be managed is exported with setting means.
2. it according to the method described in claim 1, it is characterized in that, further includes：
Learning algorithm is trained according to the abnormality detection of setting, distribution builds the abnormality detection mould of the cloud computation data center
Type.
3. according to the method described in claim 2, it is characterized in that, it is described according to the abnormality detection of setting train learning algorithm,
The abnormality detection model that distribution builds the cloud computation data center includes：
The abnormality detection training learning algorithm of setting is disposed on Distributed Computing Platform；
The history achievement data of the index to be managed is obtained from database, as abnormality detection training learning algorithm
Training dataset；
Determine the index feature of the index to be managed, the training learning characteristic as abnormality detection training learning algorithm；
Based on the training dataset and the trained learning characteristic, the training acquisition cloud computing on Distributed Computing Platform
The abnormality detection model of data center.
4. according to the method described in claim 1, it is characterized in that, obtaining the achievement data of the index to be managed includes：
Acquisition or the achievement data with index to be managed described in setting time interval acquiring in real time；
Correspondingly, the method further includes：
The achievement data of the index to be managed is stored in the database of setting, the history for forming the index to be managed refers to
Mark data.
5. according to any methods of claim 1-4, which is characterized in that the abnormality detection model based on structure determines institute
State whether achievement data includes in abnormality：
Using the achievement data as the input value of the abnormality detection model of structure, corresponding achievement data feature is generated；
If the achievement data feature meets the standard index feature of the abnormality detection model, it is determined that the achievement data
In normal condition；Otherwise, it determines the achievement data is in abnormality.
6. a kind of management system for cloud computation data center, which is characterized in that including：
Achievement data acquisition module for obtaining the index to be managed in the cloud computation data center that user determines, and obtains institute
State the achievement data of index to be managed；
Whether abnormal state detection module for the abnormality detection model based on setting, determines the achievement data in abnormal
State；
Warning information output module, for when the achievement data is in abnormality, with setting means export described in treat pipe
Manage the warning information of index.
7. system according to claim 6, which is characterized in that further include：
Detection model builds module, and for training learning algorithm according to the abnormality detection of structure, distribution builds the cloud computing
The abnormality detection model of data center.
8. system according to claim 7, which is characterized in that the detection model structure module is specifically used for：
The abnormality detection training learning algorithm of setting is disposed on Distributed Computing Platform；
The history achievement data of the index to be managed is obtained from database, as abnormality detection training learning algorithm
Training dataset；
Determine the index feature of the index to be managed, the training learning characteristic as abnormality detection training learning algorithm；
Based on the training dataset and the trained learning characteristic, the training acquisition cloud computing on Distributed Computing Platform
The abnormality detection model of data center.
9. system according to claim 6, which is characterized in that obtaining the achievement data of the index to be managed includes：
Acquisition or the achievement data with index to be managed described in setting time interval acquiring in real time；
Correspondingly, the system also includes：
History data store module for the achievement data of the index to be managed to be stored in the database of setting, is formed
The history achievement data of the index to be managed.
10. according to any systems of claim 6-9, which is characterized in that abnormal state detection module is specifically used for：
Using the achievement data as the input value of the abnormality detection model of structure, corresponding achievement data feature is generated；
If the achievement data feature meets the standard index feature of the abnormality detection model, it is determined that the achievement data
In normal condition；Otherwise, it determines the achievement data is in abnormality.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610980467.6A CN108075906A (en) | 2016-11-08 | 2016-11-08 | A kind of management method and system for cloud computation data center |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610980467.6A CN108075906A (en) | 2016-11-08 | 2016-11-08 | A kind of management method and system for cloud computation data center |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108075906A true CN108075906A (en) | 2018-05-25 |
Family
ID=62154023
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610980467.6A Pending CN108075906A (en) | 2016-11-08 | 2016-11-08 | A kind of management method and system for cloud computation data center |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108075906A (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108880881A (en) * | 2018-06-14 | 2018-11-23 | 郑州云海信息技术有限公司 | The method and apparatus of monitoring resource under a kind of cloud environment |
| CN109347978A (en) * | 2018-11-22 | 2019-02-15 | 天津众鑫锐智科技有限公司 | A kind of Intelligent internet of things computing platform |
| CN109815084A (en) * | 2018-12-29 | 2019-05-28 | 北京城市网邻信息技术有限公司 | Abnormal identification method, device and electronic device and storage medium |
| CN110413431A (en) * | 2019-08-05 | 2019-11-05 | 吉林吉大通信设计院股份有限公司 | A kind of intelligent recognition prior-warning device being directed to big data platform failure and method |
| CN110413482A (en) * | 2019-07-29 | 2019-11-05 | 中国工商银行股份有限公司 | Detection method and device |
| CN110489311A (en) * | 2019-03-01 | 2019-11-22 | 北京亿赛通科技发展有限责任公司 | Automatic detection computer user behavior and the method and system for automatically updating detection model |
| CN110740061A (en) * | 2019-10-18 | 2020-01-31 | 北京三快在线科技有限公司 | Fault early warning method and device and computer storage medium |
| WO2020078385A1 (en) * | 2018-10-18 | 2020-04-23 | 杭州海康威视数字技术股份有限公司 | Data collecting method and apparatus, and storage medium and system |
| CN111737099A (en) * | 2020-06-09 | 2020-10-02 | 国网电力科学研究院有限公司 | A Gaussian distribution-based data center anomaly detection method and device |
| CN112286771A (en) * | 2020-11-03 | 2021-01-29 | 中国电子科技集团公司第二十八研究所 | Alarm method for monitoring global resources |
| CN112486767A (en) * | 2020-11-25 | 2021-03-12 | 中移（杭州）信息技术有限公司 | Intelligent monitoring method, system, server and storage medium for cloud resources |
| CN113032242A (en) * | 2019-12-25 | 2021-06-25 | 阿里巴巴集团控股有限公司 | Data marking method and device, computer storage medium and electronic equipment |
| CN113076232A (en) * | 2021-03-30 | 2021-07-06 | 深圳供电局有限公司 | Health data index abnormity detection method and system |
| CN113138895A (en) * | 2021-04-19 | 2021-07-20 | 建信金融科技有限责任公司 | Monitoring method and device of cloud desktop equipment |
| CN113159933A (en) * | 2021-05-20 | 2021-07-23 | 中国工商银行股份有限公司 | Risk control method, system, device and medium |
| CN114328639A (en) * | 2020-09-30 | 2022-04-12 | 中强光电股份有限公司 | Abnormality detection system and abnormality detection method |
| CN115953738A (en) * | 2023-03-02 | 2023-04-11 | 上海燧原科技有限公司 | Monitoring method, device, equipment and medium for image recognition distributed training |
| CN118761699A (en) * | 2024-09-05 | 2024-10-11 | 南京汇翔自动化系统工程有限公司 | Smart material yard distribution, storage and transportation control system and method based on data analysis |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103401699A (en) * | 2013-07-18 | 2013-11-20 | 深圳先进技术研究院 | Cloud data center security monitoring early warning system and method |
| CN104794192A (en) * | 2015-04-17 | 2015-07-22 | 南京大学 | Multi-level anomaly detection method based on exponential smoothing and integrated learning model |
| CN105335816A (en) * | 2015-10-13 | 2016-02-17 | 国网安徽省电力公司铜陵供电公司 | Electric power communication operation trend and business risk analyzing method based on deep learning |
| WO2016039845A1 (en) * | 2014-09-09 | 2016-03-17 | Belkin International, Inc. | Coordinated and device-distributed detection of abnormal network device operation |
| CN105975443A (en) * | 2016-05-04 | 2016-09-28 | 西南大学 | Lasso-based anomaly detection method and system |
| CN106156026A (en) * | 2015-03-24 | 2016-11-23 | 中国人民解放军国防科学技术大学 | A kind of method based on the data online anomaly of stream fictitious assets |
-
2016
- 2016-11-08 CN CN201610980467.6A patent/CN108075906A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103401699A (en) * | 2013-07-18 | 2013-11-20 | 深圳先进技术研究院 | Cloud data center security monitoring early warning system and method |
| WO2016039845A1 (en) * | 2014-09-09 | 2016-03-17 | Belkin International, Inc. | Coordinated and device-distributed detection of abnormal network device operation |
| CN106156026A (en) * | 2015-03-24 | 2016-11-23 | 中国人民解放军国防科学技术大学 | A kind of method based on the data online anomaly of stream fictitious assets |
| CN104794192A (en) * | 2015-04-17 | 2015-07-22 | 南京大学 | Multi-level anomaly detection method based on exponential smoothing and integrated learning model |
| CN105335816A (en) * | 2015-10-13 | 2016-02-17 | 国网安徽省电力公司铜陵供电公司 | Electric power communication operation trend and business risk analyzing method based on deep learning |
| CN105975443A (en) * | 2016-05-04 | 2016-09-28 | 西南大学 | Lasso-based anomaly detection method and system |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108880881A (en) * | 2018-06-14 | 2018-11-23 | 郑州云海信息技术有限公司 | The method and apparatus of monitoring resource under a kind of cloud environment |
| CN111078488B (en) * | 2018-10-18 | 2021-11-09 | 杭州海康威视数字技术股份有限公司 | Data acquisition method, device, storage medium and system |
| WO2020078385A1 (en) * | 2018-10-18 | 2020-04-23 | 杭州海康威视数字技术股份有限公司 | Data collecting method and apparatus, and storage medium and system |
| CN111078488A (en) * | 2018-10-18 | 2020-04-28 | 杭州海康威视数字技术股份有限公司 | Data acquisition method, device, storage medium and system |
| CN109347978A (en) * | 2018-11-22 | 2019-02-15 | 天津众鑫锐智科技有限公司 | A kind of Intelligent internet of things computing platform |
| CN109815084A (en) * | 2018-12-29 | 2019-05-28 | 北京城市网邻信息技术有限公司 | Abnormal identification method, device and electronic device and storage medium |
| CN110489311A (en) * | 2019-03-01 | 2019-11-22 | 北京亿赛通科技发展有限责任公司 | Automatic detection computer user behavior and the method and system for automatically updating detection model |
| CN110489311B (en) * | 2019-03-01 | 2023-04-18 | 北京亿赛通科技发展有限责任公司 | Method and system for automatically detecting computer user behaviors and automatically updating detection model |
| CN110413482A (en) * | 2019-07-29 | 2019-11-05 | 中国工商银行股份有限公司 | Detection method and device |
| CN110413482B (en) * | 2019-07-29 | 2023-10-03 | 中国工商银行股份有限公司 | Detection methods and devices |
| CN110413431A (en) * | 2019-08-05 | 2019-11-05 | 吉林吉大通信设计院股份有限公司 | A kind of intelligent recognition prior-warning device being directed to big data platform failure and method |
| CN110740061A (en) * | 2019-10-18 | 2020-01-31 | 北京三快在线科技有限公司 | Fault early warning method and device and computer storage medium |
| CN110740061B (en) * | 2019-10-18 | 2020-09-29 | 北京三快在线科技有限公司 | Fault early warning method and device and computer storage medium |
| CN113032242A (en) * | 2019-12-25 | 2021-06-25 | 阿里巴巴集团控股有限公司 | Data marking method and device, computer storage medium and electronic equipment |
| CN113032242B (en) * | 2019-12-25 | 2024-02-09 | 阿里巴巴集团控股有限公司 | Data marking method and device, computer storage medium and electronic equipment |
| CN111737099B (en) * | 2020-06-09 | 2021-04-16 | 国网电力科学研究院有限公司 | A Gaussian distribution-based data center anomaly detection method and device |
| CN111737099A (en) * | 2020-06-09 | 2020-10-02 | 国网电力科学研究院有限公司 | A Gaussian distribution-based data center anomaly detection method and device |
| CN114328639A (en) * | 2020-09-30 | 2022-04-12 | 中强光电股份有限公司 | Abnormality detection system and abnormality detection method |
| CN112286771A (en) * | 2020-11-03 | 2021-01-29 | 中国电子科技集团公司第二十八研究所 | Alarm method for monitoring global resources |
| CN112486767A (en) * | 2020-11-25 | 2021-03-12 | 中移（杭州）信息技术有限公司 | Intelligent monitoring method, system, server and storage medium for cloud resources |
| CN113076232A (en) * | 2021-03-30 | 2021-07-06 | 深圳供电局有限公司 | Health data index abnormity detection method and system |
| CN113138895A (en) * | 2021-04-19 | 2021-07-20 | 建信金融科技有限责任公司 | Monitoring method and device of cloud desktop equipment |
| CN113159933A (en) * | 2021-05-20 | 2021-07-23 | 中国工商银行股份有限公司 | Risk control method, system, device and medium |
| CN115953738A (en) * | 2023-03-02 | 2023-04-11 | 上海燧原科技有限公司 | Monitoring method, device, equipment and medium for image recognition distributed training |
| CN118761699A (en) * | 2024-09-05 | 2024-10-11 | 南京汇翔自动化系统工程有限公司 | Smart material yard distribution, storage and transportation control system and method based on data analysis |
| CN118761699B (en) * | 2024-09-05 | 2024-11-12 | 南京汇翔自动化系统工程有限公司 | Smart material yard distribution, storage and transportation control system and method based on data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108075906A (en) | A kind of management method and system for cloud computation data center | |
| US11283863B1 (en) | Data center management using digital twins | |
| CN105989155B (en) | Identify the method and device of risk behavior | |
| EP4632530A1 (en) | Refrigeration system operation method and apparatus, device, and storage medium | |
| CN104809051B (en) | Method and apparatus for predicting exception and failure in computer application | |
| WO2020259421A1 (en) | Method and apparatus for monitoring service system | |
| US20200089533A1 (en) | Methods and systems for cloud application optimization | |
| CN112700131B (en) | AB test method and device based on artificial intelligence, computer equipment and medium | |
| US20150286949A1 (en) | Problem analysis and priority determination based on fuzzy expert systems | |
| US9530256B2 (en) | Generating cumulative wear-based indicators for vehicular components | |
| US9208209B1 (en) | Techniques for monitoring transformation techniques using control charts | |
| CN107786616A (en) | Main frame intelligent monitor system based on high in the clouds | |
| CN111897705B (en) | Service state processing and model training method, device, equipment and storage medium | |
| Alipour et al. | Online machine learning for cloud resource provisioning of microservice backend systems | |
| CN106874280A (en) | The alarm method and device of abnormal data | |
| US20240291724A1 (en) | Allocation of Resources to Process Execution in View of Anomalies | |
| CN109194534B (en) | Scheduling and management method for Internet of things equipment group | |
| CN109064104A (en) | Material equipment management method, system and terminal device | |
| CN110825549A (en) | A method, device, equipment and storage medium for determining the root cause of an information system failure | |
| CN108334427A (en) | Method for diagnosing faults in storage system and device | |
| CN113656452A (en) | Method and device for detecting abnormal index of call chain, electronic equipment and storage medium | |
| CN109726068A (en) | A kind of data detection method and device | |
| Natta | Autonomous cloud optimization leveraging AI-augmented decision frameworks | |
| CN112906727B (en) | Method and system for real-time on-line detection of virtual machine state | |
| US12348389B2 (en) | Methods and systems for cloud application optimization |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180525 |
|
| RJ01 | Rejection of invention patent application after publication |