CN109981333A - A kind of O&M method and O&M equipment applied to data center - Google Patents
A kind of O&M method and O&M equipment applied to data center Download PDFInfo
- Publication number
- CN109981333A CN109981333A CN201811622320.5A CN201811622320A CN109981333A CN 109981333 A CN109981333 A CN 109981333A CN 201811622320 A CN201811622320 A CN 201811622320A CN 109981333 A CN109981333 A CN 109981333A
- Authority
- CN
- China
- Prior art keywords
- service quality
- dimension
- evaluation value
- node
- private clound
- 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
-
- 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
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
-
- 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
- H04L43/50—Testing arrangements
- H04L43/55—Testing of service level quality, e.g. simulating service usage
-
- 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
Abstract
The embodiment of the present invention provides a kind of O&M method applied to mixing cloud data center.Mix the private clound node of cloud data center, according to private clound node after the service quality evaluation value of N number of dimension obtains comprehensive evaluation value, prediction module on the public cloud node for being deployed in mixing cloud data center sends multiple historical datas, wherein, the comprehensive evaluation value that the service quality evaluation value of N number of dimension of the private clound node obtains according to each historical data.Public cloud node receives multiple historical datas, is predicted according to comprehensive evaluation value of multiple historical datas to private clound node.This method predicts the comprehensive evaluation value of private clound node using shared node, compared to the prediction completed on private clound node, the historical data of more comprehensive evaluation values can be introduced, more massive calculating is carried out, realizes that accuracy is higher, the lower more efficient O&M of time delay.
Description
Technical field
The present invention relates to information technology fields more particularly to a kind of O&M method and O&M applied to data center to set
It is standby.
Background technique
O&M refers to through series of steps and method, management and the service of maintenance data center and/or data center
Process.Service provided by data center includes IT, the relevant service of software and internet, also includes other services.In data
The heart has been usually deployed O&M equipment.O&M equipment is used to provide O&M service to user.O&M service includes to data center
O&M, for example, data center is monitored in real time, troubleshooting, capacity management, using deployment etc..
The critical function of O&M service provided by O&M equipment first is that the monitoring of the service quality to data center, this
Outside, the very multi-functional realization of O&M equipment also relies on monitoring of the O&M equipment to the service quality of data center.Service Quality
The satisfaction for the service that amount (QoS, Quality of Service) characterization user provides data center, for measuring data
The service quality at center.The height of service quality can be measured by service quality evaluation value.O&M equipment passes through to data center
Resource monitoring so that realize monitoring to the service quality evaluation value of data center.In addition to this, it is obtained according to monitoring
The performance graph of service quality evaluation value, can also be according to the variation tendency of service quality evaluation value to service quality evaluation value
Value is predicted, to carry out the O&M of data center using the predicted value of service quality evaluation value, such as data center's event
The prediction of barrier.
Under normal conditions, prediction is carried out to the service quality evaluation value of data center and needs a large amount of computing resource and storage
Resource.When data center be mixing cloud data center, to mixing cloud data center private clound node Service Quality Metrics into
When row prediction, operand needed for the computing resource and storage resource of private clound node are often not enough to support prediction and storage are empty
Between, therefore the O&M equipment of node cannot achieve the prediction of the Service Quality Metrics to private clound node in private clound.
Summary of the invention
In a first aspect, the embodiment of the present invention provides a kind of O&M method applied to data center, which includes
Private clound node and public cloud node.Method includes: multiple history numbers that the public cloud node receives private clound node transmission
According to, the comprehensive evaluation value that the service quality evaluation value of N number of dimension of the private clound node obtains according to each historical data,
Wherein, the service quality evaluation value of N number of dimension characterizes the private clound node in the service quality of N number of dimension respectively, and N is not
Integer less than 2;The public cloud node predicts according to comprehensive evaluation value of multiple historical data to the private clound node,
Obtain predicted value;The public cloud node determines that the predicted value meets alarm regulation;Above-mentioned determination is responded, which sends
Alarm information is to the private clound node.
In O&M method provided in an embodiment of the present invention, private clound node is by sending out the historical data of comprehensive evaluation value
Send the computing capability that public cloud node is utilized to public cloud node to predict comprehensive evaluation value, thus to private clound node into
Early warning and O&M before the generation of row failure.Since public cloud node has more powerful computing capability than private clound node and deposits
Energy storage power, therefore, the embodiment of the present invention by shared node come predict private clound node comprehensive evaluation value in the way of, compared to
The prediction completed on private clound node can introduce the historical data of more comprehensive evaluation values, carry out more massive meter
It calculates.To improve the accuracy of prediction, and calculating speed is faster, provide a kind of more efficient, accurate fortune for data center
Dimension mode.
With reference to first aspect, in the first possible implementation of the first aspect, which includes being used for
The physical equipment of cloud service is provided, the service quality of N number of dimension includes the service quality of cloud service and the clothes of the physical equipment
Business quality.
The multiple service quality evaluation values for introducing different dimensions, the service provided from resource provide the work of the resource serviced
Make multiple dimensions such as state, the service quality of private clound node is investigated or monitored, so as to the fortune of private clound node
Tie up it is more accurate, can more have comprehensively react private clound node service quality.
With reference to first aspect or the first possible implementation of first aspect, second in first aspect are possible
In implementation, this method further include: the N number of dimension of the private clound node according to the private clound node in first time period
Service quality evaluation value obtain the first historical data in multiple historical data.
The comprehensive evaluation value for introducing service quality, the basis for the service quality evaluation value in the multiple dimensions of synthesis
On, comprehensive, intuitive and comprehensive parameter is provided for the service quality height of private clound node, for the clothes to data center
Quality of being engaged in more comprehensively, macroscopic view and intuitive monitoring, reduce complexity, promote user experience.
The possible implementation of second with reference to first aspect, in the third possible implementation of first aspect
In, the service quality evaluation value of the N number of dimension of the private clound node according to the private clound node in first time period obtains
The first historical data in multiple historical data, comprising: the private clound node is to N number of dimension in the first time period
Service quality evaluation value is normalized;The private clound node according to the service quality evaluation value of N number of dimension after normalization with
And the weight of the service quality evaluation value of each dimension, obtain first historical data.
The third implementation with reference to first aspect, in a fourth possible implementation of the first aspect, the party
Method further include: the private clound node obtains N* (N-1)/2 significance level parameter of the service quality evaluation value of N number of dimension,
The service quality evaluation value of any two dimension in the service quality evaluation value of each significance level parameter characterization N number of dimension
Fiducial value;The private clound node obtains the service quality evaluation of each dimension according to N* (N-1)/2 significance level parameter
The weight of value.
Second aspect, the embodiment of the present invention provide the O&M equipment of a kind of pair of data center's O&M, which is characterized in that the number
It include private clound node and public cloud node according to center, which includes: the monitoring mould being deployed on the private clound node
Block is used for: monitoring the service quality of N number of dimension of the private clound node;It is obtained according to the service quality evaluation value of N number of dimension
Comprehensive evaluation value, wherein the service quality evaluation value of N number of dimension characterizes the private clound node in the clothes of N number of dimension respectively
Business quality, N are the integer not less than 2;Multiple historical datas are sent to the prediction module being deployed on the public cloud node,
In, the comprehensive evaluation value of the service quality evaluation value acquisition of N number of dimension of the private clound node according to each historical data.
The O&M equipment further include: the prediction module being deployed on the public cloud node is used for: the multiple of monitoring modular transmission are received
Historical data;It is predicted according to comprehensive evaluation value of multiple historical data to the private clound node, obtains predicted value;It determines
The predicted value meets alarm regulation;Above-mentioned determination is responded, sends alarm information to the private clound node.
Detection module on private clound node is by being sent to the pre- of public cloud node for the historical data of comprehensive evaluation value
Module is surveyed, the computing capability of public cloud node is utilized to predict comprehensive evaluation value, to carry out failure hair to private clound node
Early warning and O&M before life.Due to comparing private clound node, public cloud node has more powerful computing capability and storage energy
Power predicts the comprehensive evaluation value of private clound node using shared node, can compared to the prediction completed on private clound node
To introduce the historical data of more comprehensive evaluation values, more massive calculating is carried out, realizes that accuracy is higher, time delay is lower
More efficient O&M.
In conjunction with second aspect, in the first possible implementation of the second aspect, which includes being used for
The physical equipment of cloud service is provided, the service quality of N number of dimension includes the service quality of cloud service and the clothes of the physical equipment
Business quality.
The multiple service quality evaluation values for introducing different dimensions, the service provided from resource provide the work of the resource serviced
Make multiple dimensions such as state, the service quality of private clound node is investigated or monitored, so as to the fortune of private clound node
Tie up it is more accurate, can more have comprehensively react private clound node service quality.
The comprehensive evaluation value for introducing service quality, the basis for the service quality evaluation value in the multiple dimensions of synthesis
On, comprehensive, intuitive and comprehensive parameter is provided for the service quality height of private clound node, for the clothes to data center
Quality of being engaged in more comprehensively, macroscopic view and intuitive monitoring, reduce complexity, promote user experience.
In conjunction with the first possible implementation of second aspect, in second of possible implementation of second aspect
In, which is used for, and is worth according to the service quality evaluation of the N number of dimension of the private clound node in first time period
To the first historical data in multiple historical data, comprising: comment the service quality of N number of dimension in the first time period
Value is normalized;It is commented according to the service quality of the service quality evaluation value of N number of dimension after normalization and each dimension
The weight of value obtains first historical data.
In conjunction with second of implementation of second aspect, in the third possible implementation of the second aspect, the prison
It surveys module to be also used to: obtaining N* (N-1)/2 significance level parameter of the service quality evaluation value of N number of dimension, it is each important
Extent index characterizes the comparison of the service quality evaluation value of any two dimension in the service quality evaluation value of N number of dimension
Value;According to N* (N-1)/2 significance level parameter, the weight of the service quality evaluation value of each dimension is obtained.
The third aspect, the embodiment of the present invention provide a kind of data center, which is characterized in that the data center includes at least one
A calculating equipment, which includes processor and memory, which executes the program in the memory
Instruction, to realize the various methods of public cloud node and the execution of private clound node in first aspect.
Fourth aspect, the embodiment of the present invention provide a kind of computer program product and non-volatile computer readable storage medium
Matter includes wherein computer instruction in computer program product and non-volatile computer readable storage medium storing program for executing, calculates equipment and hold
Row computer instruction is for realizing the various methods in first aspect of the embodiment of the present invention.
Detailed description of the invention
Fig. 1 is a kind of schematic diagram of data center architecture provided in an embodiment of the present invention;
Fig. 2 is a kind of schematic diagram for mixing cloud data center provided in an embodiment of the present invention;
Fig. 3 is a kind of schematic diagram of O&M deployed with devices provided in an embodiment of the present invention
Fig. 4 is a kind of schematic diagram of the method for data center's O&M in the embodiment of the present invention;
Fig. 5 is a kind of schematic diagram for the method for obtaining comprehensive evaluation value in the embodiment of the present invention;
Fig. 6 is a kind of schematic diagram of O&M equipment in the embodiment of the present invention;
Fig. 7 is a kind of schematic diagram of the calculating equipment in the embodiment of the present invention in data center.
Specific embodiment
Shown in the data center 100 of data center as shown in figure 1 in the embodiment of the present invention.Data center 100 includes resource
110, it is based on resource 110, data center 100 provides service 120.Service 120 is deployed in resource 110.Service 120 includes fortune
Dimension service 121, calculating service, storage service, network service, management service, data service, security service etc..O&M service 121
For operation/maintenance data center 100.Resource 110 includes physical resource and/or virtual resource, and specific to drip, resource 110 includes calculating money
Source 111, storage resource 112, Internet resources 113 and O&M equipment 140 etc..Computing resource 111 includes for being data center 100
The calculating equipment of computing capability, including physical computing devices and/or virtual computing device are provided, for example, physical server, or fortune
Virtual machine or container of the row on physical server.Storage resource 112 includes for providing storage capacity for data center 100
Store equipment, including physical storage device and/or virtual depositor's equipment, such as storage array or virtual memory facilities.Network
Resource 113 includes including the physical network device and/or virtual for providing the network equipment of storage capacity for data center 100
Network equipment, such as interchanger, router, virtual switch, virtual router etc..In practical application, computing resource 111 is deposited
Storage resource 112 and Internet resources 113 can be deployed in data center 100.Computing resource 111, storage resource 112, network money
Calculating equipment, storage equipment and the network equipment in source 113, can be used for directly providing service for user, it can also be used to support
Or management is supplied to the service etc. of user.
The data center 100 for being deployed with virtual machine, virtual memory facilities or virtual network device is cloud data center.
Cloud data center is based on resource 110, provides a user cloud service on demand, the resource 110 of cloud data center include physical resource and
Virtual resource.
Cloud data center includes publicly-owned cloud data center, privately owned cloud data center and mixing cloud data center.
Publicly-owned cloud data center is the cloud environment used by several tissues and/or user sharing.In publicly-owned cloud data center
In, service needed for user is independent by one, third party provider provides, this publicly-owned cloud data center of all user sharings
On all resources.
Privately owned cloud data center is the data center exclusively enjoyed by some tissue or user.The public affairs provided by third party provider
Having cloud data center usually has powerful computing capability and storage capacity.In privately owned cloud data center, if data center is
Some tissue exclusively enjoys, and the member of tissue shares all resources of the privately owned cloud data center, be not belonging to the user of the tissue without
Method accesses the service of this data center offer;If data center is that some user exclusively enjoys, other users can not access this
The service that a data center provides.Under normal circumstances, the computing capability of privately owned cloud data center and storage capacity are weaker than publicly-owned
Cloud data, but due to privately owned cloud data center by tissue or user exclusively enjoy, the safety of privately owned cloud data center is higher.
The advantages of mixing cloud data center combines both publicly-owned cloud data center and privately owned cloud data center.Such as Fig. 2 institute
Show, mixing cloud data center 200 includes public cloud node 212 and private clound node 211.Public cloud node 212 and private clound section
Point 211 all has computing resource, storage resource and Internet resources.The service 120 for mixing cloud data center 200 is based on public cloud section
Point 212 and private clound node 211 are disposed, and service 120 includes O&M service 121.Public cloud node 212 has powerful calculating energy
Power and storage capacity, by several tissues of its resource and/or user sharing;The resource of private clound node 211 is by some tissue or uses
Family exclusively enjoys, to provide higher security performance for the tissue or user.Mix the public cloud node 212 of cloud data center 200
The service of upper deployment generally requires powerful calculating ability or storage capacity, but the requirement to security performance is relatively low;And it disposes
In the service of private clound node 211, computing capability or storage capacity are required lower but more demanding to security performance.
The embodiment of the present invention utilizes public cloud node during to the privately owned node O&M in cloud data center is mixed
Computing capability, the Service Quality Metrics of private clound node are predicted.
The embodiment of the invention provides a kind of O&M methods of data center.This method can be applied in mixed cloud data
The heart 200, for providing the O&M service 121 to mixing cloud data center 200.This method can O&M equipment as shown in Figure 3
300 execute.As shown in figure 3, O&M equipment 300 is deployed in mixing cloud data center 200.Specifically, O&M equipment 300 is wrapped
Include the first O&M unit 310 and the second O&M unit 320;First O&M unit 310 is deployed in private clound node 211, by private
There are computing resource, storage resource and Internet resources in cloud node 211 to realize;Second O&M unit 320 is deployed in public cloud section
In point 212, realized by computing resource, storage resource and the Internet resources in public cloud node 211.Below in conjunction with Fig. 3 and Fig. 4
The O&M method of description of the embodiment of the present invention is described.As shown in figure 4, this method includes the following steps.
S401, the first O&M unit 310 of private clound node 211 obtain the Service Quality of N number of dimension of private clound node 211
The multiple groups historical data of evaluation of estimate is measured, the service quality evaluation value of N number of dimension characterizes private clound node 211 in N number of dimension respectively
Service quality height, N is integer not less than 2, and every group of historical data includes the service of N number of dimension in a period
Quality evaluation value.
Illustratively, partial service quality evaluation is shown in table 1, is belonging respectively to private clound service quality, service
Device service quality, storage service quality, network service quality, the respectively different types of service such as performance, availability, reliability
Quality evaluation.Each service quality evaluation is service quality of the characterize data center under corresponding dimension, for example, privately owned cloud service
This service quality evaluation of response time is performance indicator, it characterizes private clound node in privately owned cloud service to the sound of service request
Answer the service quality of this dimension of speed.N number of service quality evaluation in the embodiment of the present invention is not limited to service shown in table 1
Quality index.
Table 1
The first O&M unit 310 of s402, private clound node 211 obtain the more of comprehensive index value according to multiple groups historical data
A historical data, each historical data are comprehensive to be obtained according to the service quality evaluation value of N number of dimension of private clound node 211
Close evaluation of estimate.
Specifically, the first O&M unit 310 of private clound node 211 is according to private clound node 211 in first time period
The service quality evaluation value of N number of dimension obtain the first historical data in multiple historical datas of comprehensive index value, wherein
One historical data is one in multiple historical datas of comprehensive index value, and first time period is that the service quality of N number of dimension is commented
One in the multiple groups historical data of value corresponding multiple periods.
Under normal conditions, after one group of historical data for obtaining the service quality evaluation value of N number of dimension of a period, i.e.,
It is calculated according to historical data of this group of historical data to the comprehensive evaluation value of the period.The embodiment of the present invention provides one kind
A historical data method for obtaining comprehensive evaluation value according to service quality evaluation value in a period is as shown in Figure 5.
The service quality evaluation value of N number of dimension is normalized in s4021.
The service quality evaluation value of N number of dimension may have different units, such as storage average time between failures, object
Manage server draw time between failures unit be the second, and store equipment availability, the unit of physical server availability is
It is secondary.Before obtaining comprehensive evaluation value, need that the service quality evaluation value of N number of dimension is normalized to eliminate its unit.
Specifically, the embodiment of the present invention provides what a kind of service quality evaluation value to N number of dimension was normalized
Formula.According to the following formula to service quality evaluation value xiIt is normalized, obtains normalized service quality evaluation value
yi:
Wherein, the arbitrary integer that the value of i is 1 to N, xiFor any service in the service quality evaluation value of N number of dimension
Quality evaluation value, yiFor the service quality evaluation value after normalization, min is the smallest in the service quality evaluation value of N number of dimension
Service quality evaluation value, max are maximum service quality evaluation value in the service quality evaluation value of N number of dimension.
S4022, using the thought of multiple attribute decision making (MADM) (Multiple Attribute Decision Making, MADM),
According to the weight of each Service Quality Metrics, the service quality evaluation value of N number of dimension after processing normalization obtains overall merit
Value.Service quality evaluation value xiWeight wiThe service quality evaluation value is characterized in the service quality evaluation value according to N number of dimension
Importance when height to evaluate the service quality of data center.Specifically, comprehensive evaluation value P is obtained according to the following formula:
In the case where being not easy to obtain the weight of each service quality evaluation, the embodiment of the invention provides a kind of according to weight
The method for wanting extent index to obtain the weight of each service quality evaluation value.
The ratio of any two service quality evaluation value in the service quality evaluation value of the N number of dimension of significance level parameter characterization
Compared with value.The Service Quality Metrics evaluation of estimate of N number of dimension corresponds to N* (N-1)/2 significance level parameter, important with N* (N-1)/2
Extent index be matrix element, Judgement Matricies A, then the corresponding characteristic vector W of the Maximum characteristic root of judgment matrix A be
Characterize the weight of the service quality evaluation value of N number of dimension.
Wherein, aijFor important extent index, the value of i, j are 1 to N integer, aijCharacterize xiCorresponding service quality
Evaluation of estimate and xjThe fiducial value of corresponding service quality evaluation value.Characteristic vector W is 1*N rank matrix, and the member of feature vector is
For the weight of the service quality evaluation value of N number of dimension, i.e. W=(w1, w2..., wn), wiService Quality Metrics xiWeight.
Under normal conditions, the first O&M unit of private clound node can monitor the service quality evaluation value of N number of dimension in real time,
And comprehensive evaluation value is calculated according to the value of the service quality evaluation value of N number of dimension in real time.Due in private clound storage resource,
Computing resource is limited, and after the historical data for obtaining comprehensive evaluation value, the first O&M unit is by the historical data of comprehensive evaluation value
It is uploaded to public cloud node, the historical data of comprehensive evaluation value is stored in public cloud node.
S403, the second O&M unit 320 of public cloud node 212 obtain the comprehensive evaluation value that the private clound node is sent
Multiple historical datas.
S404, public cloud node 212 is based on multiple historical datas of comprehensive evaluation value to the fortune of the service of private clound node
Row situation is predicted.
Specifically, the second O&M unit obtains overall merit using multiple historical datas of comprehensive evaluation value as training set
The prediction model of value.It, can be according to multiple historical datas comprising comprehensive evaluation value by neural network and the method for deep learning
Training set, obtain the prediction model of comprehensive evaluation value.Preferably, training method includes recurrent neural network (Recurrent
Neural Network, RNN) coaching method, especially shot and long term memory (Long Short-Term Memory, LSTM) training
Method.In addition to this, any method for obtaining prediction model according to training set can be used in the embodiment of the present invention.
Second O&M unit is based on prediction model and predicts comprehensive evaluation value, obtains predicted value.Predicted value reflects
The operation conditions trend of the service of private clound node.
S405, the second O&M unit 320 of public cloud node 212 determine that predicted value meets alarm regulation.
S405, responds above-mentioned determination, and alarm information is sent to privately owned by the second O&M unit 320 of public cloud node 212
First O&M unit 310 of cloud node 211 so that the first O&M unit 310 according to alarm information to private clound node 211 into
Row O&M.
S406, the first O&M unit 310 of private clound node 211 is according to the alarm information received, to private clound node
211 carry out O&M, such as fault inquiry, troubleshooting, dilatation.
The overall target of data center can be obtained by the above method, to realize straight to the service quality of data center
It sees, comprehensive, quantitative evaluation, improves data center's O&M efficiency, convenient for the operation of the O&Ms such as subsequent early warning, fault identification.
O&M equipment 300 in the embodiment of the present invention includes the first O&M unit 310 and the second O&M unit 320.Such as Fig. 6
Shown, which includes private clound node and public cloud node.First O&M unit 310 include monitoring modular 311,
Processing module 312;Second O&M module includes including prediction module 313.Each module on first O&M unit 310 is disposed respectively
Each module on private clound node 211, the second O&M unit 320 is deployed in public cloud node 313 respectively.
Monitoring modular 311, is used for: the service quality of N number of dimension of monitoring private clound node 211;According to N number of dimension
Service quality evaluation value obtains comprehensive evaluation value, wherein the service quality evaluation value of N number of dimension characterizes private clound node respectively
211 N number of dimension service quality, N is integer not less than 2;To the prediction module being deployed on public cloud node 212
313 send multiple historical datas, wherein each historical data is the service quality according to N number of dimension of private clound node 211
The comprehensive evaluation value that evaluation of estimate obtains.
Prediction module 313, is used for: receiving multiple historical datas that monitoring modular 311 is sent;According to multiple historical data
The comprehensive evaluation value of private clound node 211 is predicted, predicted value is obtained;Determine that the predicted value meets alarm regulation;Response
Above-mentioned determination sends the processing module 312 of alarm information to the private clound node 211.
312 song fan's alarm information of processing module carries out O&M to private clound node 211.
Optionally, private clound node 211 includes for providing the physical equipment of service 120 as shown in Figure 2, this is N number of
The service quality of dimension includes the service quality for servicing 120 and the service quality of the physical equipment.
Optionally, monitoring modular 311 is used for, according to the clothes of the N number of dimension of the private clound node 211 in first time period
Business quality evaluation value obtains the first historical data in multiple historical data, comprising: to N number of dimension in the first time period
The service quality evaluation value of degree is normalized;According to the service quality evaluation value and each dimension of N number of dimension after normalization
The weight of the service quality evaluation value of degree obtains first historical data.
Optionally, which is also used to: obtaining the N* (N-1)/2 of the service quality evaluation value of N number of dimension
Significance level parameter, any two dimension in the service quality evaluation value of each significance level parameter characterization N number of dimension
The fiducial value of service quality evaluation value;According to N* (N-1)/2 significance level parameter, the service quality for obtaining each dimension is commented
The weight of value.
It is as shown in Figure 7 that the embodiment of the present application also provides a kind of data center 700.Data center 700 includes at least one meter
It calculates equipment 710 and at least one calculates equipment 720.Data center 700 can be used for realizing in mixed cloud data as shown in Figure 3
The heart 200, mix the public cloud node 212 in cloud data center 200, private clound node 211, O&M equipment 300 be deployed in
In few calculating equipment 710 and/or at least one calculating equipment 720.Specifically, private clound node 211 is deployed at least one
In a calculating equipment 710, public cloud node 212 is deployed at least one and calculates in equipment 720.Accordingly, on private cloud node 211
The first O&M unit 310 be deployed at least one calculate equipment 710 on, the second O&M unit 320 on public cloud node 212
At least one is deployed in calculate in equipment 720.Calculating equipment 710 may include processing unit 711 and communication interface 712, processing
Unit 711 is used to execute function defined in the operating system and various software programs for calculating and running in equipment, including aforementioned
The function of each module in first O&M unit 310.Calculating equipment 720 may include processing unit 721 and communication interface 722, place
Reason unit 721 is used to execute function defined in the operating system and various software programs for calculating and running in equipment, including preceding
State the function of each module in the second O&M unit 320.Communication interface 712 and communication interface 722 are used to carry out with other equipment
Communication interaction, other equipment can be other calculating equipment, and specifically, communication interface 712 and communication interface 722 can be net
Network adapter.
Optionally, calculating equipment 710 can also include input/output interface 713, and input/output interface 713 is connected with defeated
Enter/output equipment, information for receiving input exports operating result.Input/output interface 713 can for mouse, keyboard,
Display or CD-ROM drive etc..Optionally, which can also include additional storage 714, also commonly referred to as external memory,
The storage medium of additional storage 714 can be magnetic medium (for example, floppy disk, hard disk, tape), optical medium (such as CD),
Or semiconductor medium (such as solid state hard disk) etc..Processing unit 711 can there are many specific implementation forms, such as processing unit
711 may include processor 7111 and memory 7112, and processor 7111 executes phase according to the program instruction stored in memory 7112
The operation of pass, processor 7111 can be central processing unit (CPU) or image processor (graphics processing
Unit, GPU), processor 7111 can be single core processor or multi-core processor.Processing unit 711 can also be individually using interior
The logical device of processing logic is set to realize, such as field programmable gate array (full name in English: Field Programmable
Gate Array, abbreviation: FPGA) or digital signal processor (English: digital signal processor, DSP) etc..This
Outside, the calculating equipment 710 in Fig. 7 is only an example of a calculating equipment, and calculating equipment 710 may be comprising compared to figure
More perhaps less component shown in 7 has different component Configuration modes.
Similarly, calculating equipment 720 also may include the processing unit 712 that input/output interface 713. calculates equipment 720
Also a variety of specific implementation forms be can have, such as processing unit 721 may include processor 7211 and memory 7212, processor
7211 execute relevant operation according to the program instruction stored in memory 722, or individually using the logic of built-in processing logic
Device is realized.Calculating equipment 720 may include compared to the more perhaps less component for calculating equipment 710 or to have not
Same component Configuration mode.
In several embodiments provided herein, it should be understood that disclosed systems, devices and methods, it can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
It divides, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components
It can be combined or can be integrated into another system, or some features can be ignored or not executed.In addition, shown or beg for
Opinion mutual coupling, direct-coupling or communication connection can be through some interfaces, the INDIRECT COUPLING of device or unit
Or communication connection, it is also possible to electricity, mechanical or other form connections.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can readily occur in various equivalent modifications or replace
It changes, these modifications or substitutions should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with right
It is required that protection scope subject to.
Claims (10)
1. a kind of O&M method applied to data center, which is characterized in that the data center includes private clound node and public affairs
There is Yun Jiedian, which comprises
The public cloud node receives multiple historical datas that the private clound node is sent, each historical data is according to institute
State the comprehensive evaluation value that the service quality evaluation value of N number of dimension of private clound node obtains, wherein the service of N number of dimension
Quality evaluation value characterizes the private clound node in the service quality of N number of dimension respectively, and N is the integer not less than 2;
The public cloud node is predicted according to comprehensive evaluation value of the multiple historical data to the private clound node, is obtained
To predicted value;
The public cloud node determines that the predicted value meets alarm regulation;
Above-mentioned determination is responded, the public cloud node sends alarm information to the private clound node.
2. method according to claim 1, which is characterized in that the private clound node includes for providing cloud service
Physical equipment, the service quality of N number of dimension include the service quality of cloud service and the service quality of the physical equipment.
3. according to claim 1 or method described in 2, which is characterized in that the method also includes:
The service quality evaluation of the N number of dimension of the private clound node according to the private clound node in first time period
Value obtains the first historical data in the multiple historical data.
4. method according to claim 3, which is characterized in that the private clound node exists according to the private clound node
The service quality evaluation value of N number of dimension in first time period obtains the first history number in the multiple historical data
According to, comprising:
The service quality evaluation value of N number of dimension in the first time period is normalized in the private clound node;
The private clound node is according to the service quality evaluation value of N number of dimension after normalization and the service quality of each dimension
The weight of evaluation of estimate obtains first historical data.
5. method according to claim 3, which is characterized in that the method also includes:
The private clound node obtains N* (N-1)/2 significance level parameter of the service quality evaluation value of N number of dimension, often
The service quality evaluation value of any two dimension in the service quality evaluation value of N number of dimension described in a significance level parameter characterization
Fiducial value;
The private clound node obtains the service quality evaluation of each dimension according to the N* (N-1)/2 significance level parameter
The weight of value.
6. the O&M equipment of a kind of pair of data center's O&M, which is characterized in that the data center includes private clound node and public affairs
There is Yun Jiedian, the O&M equipment includes:
The monitoring modular being deployed on the private clound node, is used for:
Monitor the service quality of N number of dimension of the private clound node;
Comprehensive evaluation value is obtained according to the service quality evaluation value of N number of dimension, wherein the service quality of N number of dimension
Evaluation of estimate characterizes the private clound node in the service quality of N number of dimension respectively, and N is the integer not less than 2;
Multiple historical datas are sent to the prediction module that is deployed on the public cloud node, wherein each historical data is
The comprehensive evaluation value obtained according to the service quality evaluation value of N number of dimension of the private clound node;
The prediction module, is used for:
Receive the multiple historical data that the monitoring modular is sent;
It is predicted according to comprehensive evaluation value of the multiple historical data to the private clound node, obtains predicted value;
Determine that the predicted value meets alarm regulation;
Above-mentioned determination is responded, sends alarm information to the private clound node.
7. O&M equipment according to claim 1, which is characterized in that the private clound node includes for providing cloud clothes
The physical equipment of business, the service quality of N number of dimension include the service quality of cloud service and the Service Quality of the physical equipment
Amount.
8. O&M equipment according to claim 6 or 7, which is characterized in that the monitoring modular is used for:
The service quality evaluation value of N number of dimension in first time period is normalized;
According to the service quality evaluation value of N number of dimension after normalization and the weight of the service quality evaluation value of each dimension,
Obtain the first historical data in the multiple historical data.
9. O&M equipment according to claim 8, which is characterized in that the monitoring modular is also used to:
Obtain N* (N-1)/2 significance level parameter of the service quality evaluation value of N number of dimension, each significance level parameter
Characterize the fiducial value of the service quality evaluation value of any two dimension in the service quality evaluation value of N number of dimension;
According to the N* (N-1)/2 significance level parameter, the weight of the service quality evaluation value of each dimension is obtained.
10. a kind of data center, which is characterized in that the data center includes that the first calculating equipment and second calculate equipment, institute
Stating the first calculating equipment includes first processor and first memory, and the second calculating equipment includes second processor and second
Memory, the first processor execute the program instruction in the first memory, to realize described in claim 1-5
The method that private clound node executes, the second processor executes the program instruction in the second memory, to realize right
It is required that the method that public cloud node described in 1-5 executes.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811622320.5A CN109981333B (en) | 2018-12-28 | 2018-12-28 | Operation and maintenance method and operation and maintenance equipment applied to data center |
PCT/CN2019/129603 WO2020135806A1 (en) | 2018-12-28 | 2019-12-28 | Operation maintenance method and equipment applied to data center |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811622320.5A CN109981333B (en) | 2018-12-28 | 2018-12-28 | Operation and maintenance method and operation and maintenance equipment applied to data center |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109981333A true CN109981333A (en) | 2019-07-05 |
CN109981333B CN109981333B (en) | 2022-03-25 |
Family
ID=67076482
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811622320.5A Active CN109981333B (en) | 2018-12-28 | 2018-12-28 | Operation and maintenance method and operation and maintenance equipment applied to data center |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN109981333B (en) |
WO (1) | WO2020135806A1 (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020135806A1 (en) * | 2018-12-28 | 2020-07-02 | 华为技术有限公司 | Operation maintenance method and equipment applied to data center |
CN111416735A (en) * | 2020-03-02 | 2020-07-14 | 河海大学 | Federal learning-based safety QoS prediction method under mobile edge environment |
CN113590571A (en) * | 2021-09-29 | 2021-11-02 | 睿至科技集团有限公司 | Method and system for sharing private cloud resources and public cloud resources |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112118131B (en) * | 2020-09-01 | 2023-07-25 | 紫光云(南京)数字技术有限公司 | Cloud resource management method with high reliability and high speed capacity expansion |
CN112561318A (en) * | 2020-12-14 | 2021-03-26 | 清华大学 | Comprehensive evaluation and analysis tool for energy system of data center |
CN112667594A (en) * | 2021-01-14 | 2021-04-16 | 北京智源人工智能研究院 | Heterogeneous computing platform based on hybrid cloud resources and model training method |
CN116614431B (en) * | 2023-07-19 | 2023-10-03 | 中国电信股份有限公司 | Data processing method, device, electronic equipment and computer readable storage medium |
CN117033880B (en) * | 2023-10-10 | 2024-01-05 | 北京金信润天信息技术股份有限公司 | Automatic operation and maintenance method, device, equipment and storage medium for data center |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140351648A1 (en) * | 2013-05-24 | 2014-11-27 | Connectloud, Inc. | Method and Apparatus for Dynamic Correlation of Large Cloud Firewall Fault Event Stream |
CN106886469A (en) * | 2017-04-10 | 2017-06-23 | 深圳第线通信有限公司 | A kind of cloud computing disaster tolerance management method |
CN107895176A (en) * | 2017-11-13 | 2018-04-10 | 国网湖南省电力有限公司 | A kind of mist computing system and method towards water power group of planes wide area monitoring, diagnosing |
CN107992951A (en) * | 2017-12-11 | 2018-05-04 | 上海市信息网络有限公司 | Capacity alarm method, system, memory and the electronic equipment of cloud management platform |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10225164B2 (en) * | 2012-09-07 | 2019-03-05 | Oracle International Corporation | System and method for providing a cloud computing environment |
CN106293872A (en) * | 2016-07-27 | 2017-01-04 | 云南电网有限责任公司信息中心 | A kind of SLA resources balance management-control method based on resource pool |
CN109981333B (en) * | 2018-12-28 | 2022-03-25 | 华为云计算技术有限公司 | Operation and maintenance method and operation and maintenance equipment applied to data center |
-
2018
- 2018-12-28 CN CN201811622320.5A patent/CN109981333B/en active Active
-
2019
- 2019-12-28 WO PCT/CN2019/129603 patent/WO2020135806A1/en active Application Filing
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140351648A1 (en) * | 2013-05-24 | 2014-11-27 | Connectloud, Inc. | Method and Apparatus for Dynamic Correlation of Large Cloud Firewall Fault Event Stream |
CN106886469A (en) * | 2017-04-10 | 2017-06-23 | 深圳第线通信有限公司 | A kind of cloud computing disaster tolerance management method |
CN107895176A (en) * | 2017-11-13 | 2018-04-10 | 国网湖南省电力有限公司 | A kind of mist computing system and method towards water power group of planes wide area monitoring, diagnosing |
CN107992951A (en) * | 2017-12-11 | 2018-05-04 | 上海市信息网络有限公司 | Capacity alarm method, system, memory and the electronic equipment of cloud management platform |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2020135806A1 (en) * | 2018-12-28 | 2020-07-02 | 华为技术有限公司 | Operation maintenance method and equipment applied to data center |
CN111416735A (en) * | 2020-03-02 | 2020-07-14 | 河海大学 | Federal learning-based safety QoS prediction method under mobile edge environment |
CN111416735B (en) * | 2020-03-02 | 2021-05-11 | 河海大学 | Federal learning-based safety QoS prediction method under mobile edge environment |
CN113590571A (en) * | 2021-09-29 | 2021-11-02 | 睿至科技集团有限公司 | Method and system for sharing private cloud resources and public cloud resources |
CN113590571B (en) * | 2021-09-29 | 2022-01-18 | 睿至科技集团有限公司 | Method and system for sharing private cloud resources and public cloud resources |
Also Published As
Publication number | Publication date |
---|---|
WO2020135806A1 (en) | 2020-07-02 |
CN109981333B (en) | 2022-03-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109981333A (en) | A kind of O&M method and O&M equipment applied to data center | |
US10841241B2 (en) | Intelligent placement within a data center | |
US11582130B2 (en) | Performance monitoring in a distributed storage system | |
CN105279027B (en) | A kind of virtual machine deployment method and device | |
US8756441B1 (en) | Data center energy manager for monitoring power usage in a data storage environment having a power monitor and a monitor module for correlating associative information associated with power consumption | |
US11206193B2 (en) | Method and system for provisioning resources in cloud computing | |
CN103763130B (en) | Management method, the device and system of large-scale cluster | |
CN106170947B (en) | A kind of alarm information processing method, relevant device and system | |
US11275617B2 (en) | Self-managed intelligent elastic cloud stack | |
WO2012033909A2 (en) | Method and system of simulating a data center | |
Guo et al. | Scaling exact multi-objective combinatorial optimization by parallelization | |
CN105765556A (en) | Customer-directed networking limits in distributed systems | |
CN108092813A (en) | Data center's total management system server hardware Governance framework and implementation method | |
Ataie et al. | Modeling and evaluation of dispatching policies in IaaS cloud data centers using SANs | |
Smith | A system for monitoring and management of computational grids | |
CN109710447A (en) | For the method, apparatus of data access, medium and calculate equipment | |
US9547747B2 (en) | Distributed internet protocol network analysis model with real time response performance | |
CN109274533A (en) | A kind of positioning device and method of the Web service failure of rule-based engine | |
CN106227601A (en) | Method and apparatus for the resource distribution of virtual platform | |
Reynolds et al. | Provisioning norm: An asymmetric quality measure for saas resource allocation | |
Hardwick et al. | Modeling the performance of e-commerce sites | |
Naskos et al. | Elton: a cloud resource scaling-out manager for nosql databases | |
Zhang et al. | A comprehensive toolset for workload characterization, performance modeling, and online control | |
Hyder et al. | Power Aware Virtual Machine Migration for Resource Allocation in Cloud | |
Kim et al. | A Multi service capacitated facility location problem with stochastic demand |
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 | ||
TA01 | Transfer of patent application right | ||
TA01 | Transfer of patent application right |
Effective date of registration: 20220225 Address after: 550025 Huawei cloud data center, jiaoxinggong Road, Qianzhong Avenue, Gui'an New District, Guiyang City, Guizhou Province Applicant after: Huawei Cloud Computing Technology Co.,Ltd. Address before: 518129 Bantian HUAWEI headquarters office building, Longgang District, Guangdong, Shenzhen Applicant before: HUAWEI TECHNOLOGIES Co.,Ltd. |
|
GR01 | Patent grant | ||
GR01 | Patent grant |