CN104111875B - Cloud data center increases number of tasks device for controlling dynamically, system and method newly - Google Patents
Cloud data center increases number of tasks device for controlling dynamically, system and method newly Download PDFInfo
- Publication number
- CN104111875B CN104111875B CN201410315765.4A CN201410315765A CN104111875B CN 104111875 B CN104111875 B CN 104111875B CN 201410315765 A CN201410315765 A CN 201410315765A CN 104111875 B CN104111875 B CN 104111875B
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- data center
- mtd
- newly
- 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.)
- Active
Links
Abstract
The invention discloses a kind of cloud data center to increase number of tasks device for controlling dynamically, system and method newly, belong to field of cloud calculation, the variation tendency of load and physical machine resource when the present invention is by tracking cloud data center actual motion, then according to the predicted value that newly-increased task amount can be born to system in future, it is dynamic to determine new task window size, avoid Task Congestion and computing resource from dallying, reach optimization balance, taken into account the performance and economy of cloud system.
Description
Technical field
The invention belongs to field of cloud calculation, more particularly to a kind of cloud data center increase newly number of tasks device for controlling dynamically,
System and method.
Background technology
Cloud computing refers to by task distribution on substantial amounts of distributed computer, is using cloud computing platform, by network
User provides the computation schema of information service.Relative to traditional software forms, cloud computing have loose couplings, on-demand,
The significant advantages such as cost is controllable, resource is virtual, isomery collaboration, make it more adapt to ecommerce, flexible manufacturing, movement now
Apply internet etc..Cloud computing includes the implication of two aspects：It is the cloud computing platform infrastructure of bottom structure on one side,
It is the basis for building upper level applications；Another aspect is meant that cloud computing of the structure on this basic platform should
Use program.
It is data center that cloud computing, which carries out entity computing resource and the form of computing unit integration,.Data center is to pass through to be
Physical resource is virtualized into money by the virtualization technologies such as system virtualization, multiprocessor virtualization, internal memory virtualization, I/O virtualizations
Source pond, these resources are planned as a whole by the management and dispatching module of the software and hardware in data center again.
Main computing resource in the resource pool of cloud data center, is exactly physical machine (Physical Machine, PM), thing
Reason machine is that (i.e. a physical machine can not split into more sub- things again to the most basic cloud task executing units of can not subdivide one
Reason machine), it is a certain amount of calculating, storage and set of networked communication resource.One or more can be run in one physical machine
Virtual machine process, and a virtual machine process synchronization is only possible to run in a physical machine.
In cloud data center running, the physical machine in resource pool is always at the dynamic that task is received, performs, discharged
In change.After the new task that user submits to data center, the task management of data center and Dispatching Unit will be acceptable
New task is translated as virtual machine process (VM process), and is distributed in the physical machine of non-full load and performs.
Traditional cloud data center task management technology, always set the newly-increased acceptable task window of a constant size
Mouthful, the shortcomings that this technology is present：The dynamic change situation of physical machine load is not accounted for, though easily have in physical machine surplus
Still arrange more newly-increased task in the case of remaining disposal ability but the very fast growth of load, cause task distribution one section
Occurs Task Congestion after time, systematic function drastically declines；In turn, it is though also easily very fast in the higher load of physical machine load
Still less newly-increased task is arranged in the case of decline, cause task distribution a period of time after there is physical machine idle running
Situation, cause computing resource waste.
The content of the invention
In view of the drawbacks described above of prior art, the technical problems to be solved by the invention are to provide a kind of optimization cloud data
The newly-increased number of tasks dynamic control method of the efficiency of central task management of performance and managing computing resources.
To achieve the above object, the invention provides a kind of cloud data center to increase number of tasks dynamic control method newly, according to
Interval time t carries out task distribution to cloud data center, carries out according to the following steps：
Step 1: obtain system information；The system information includes：The newly-increased task quantity XZT of epicycle data centerx、
The task quantity WCT that epicycle data center completesx, in data center the task that each physical machine is currently run quantity D QTiWith
The task immigration number QYT that each physical machine occurs in epicyclei；Wherein 1≤i≤WLJ, WLJ are physical machine in data center
Quantity；
Step 2: being analyzed the system information of acquisition and decision-making；
The system information of described pair of acquisition is analyzed and decision-making is carried out according to the following steps：
A1, data prediction is carried out to the system information got；
Calculate the tasks carrying rate ZXL of data center's epicyclex, calculate each physical machine and can receive new task and account for data center
The ratio BL of overall newly-increased number of tasksi, calculate ZXL1To ZXLxThe step-length average increment INC for eliminating excessive value interference of sequence,
Calculate expected future cloud data center tasks carrying rate FZXL, calculate the expected future execution of each physical machine in cloud data center
Rate FZXLPMi；
Calculate the tasks carrying rate ZXL of data center's epicyclex：
Mean { } is the operation that set is averaging, and x is epicycle number, 1≤x<Between ∞, t are between often wheel set in advance
Every the time；
The ratio BL that new task accounts for the newly-increased number of tasks of data center's totality can be received by calculating each physical machinei：
Calculate ZXL1To ZXLxThe step-length average increment INC for eliminating excessive value interference of sequence：
The YCu,vFor judge whether be excessive value token variable, ZLu,vRepresent in sequence between u and v record value
The equivalent increment of power series of formation；U and v is positive integer；
Next expected future cloud data center tasks carrying rate FZXL is calculated：
The γ is the distance reduction factor, and its effect is nearest historical record value is had a great influence FZXL, and is got over
Early influence is smaller；γ meets 0.5<γ<1；
Calculate the expected future implementation rate FZXLPM of each physical machine in cloud data centeri：
FZXLPMi=FZXL × BLi；
A2, calculate control decision reference value；
CalculateObtain each physical machine future it is pre-
Phase remaining acceptable number of tasks FZXLPM；The ZDT is the supported maximum task quantity of single physical machine；
Cloud data center next round is calculated to be expected newly to enter number of tasks KXZT：
A3, send decision information；
Calculate cloud data center next stage newly-increased task window controlled quentity controlled variable KZL：
Step 3: newly-increased task management：If KZL is 0, in next round, cloud data center refuses all newly-increased tasks；If
KZL is more than 0, then in next round, cloud data center receives first to KZL newly-increased task, and refuses the institute after KZL
There is newly-increased task；Wheel number x is returned after adding 1 and is performed step 1 (i.e. x=x+1).
Using above technical scheme, loaded according to the real time execution of each physical machine, dynamically determine that next stage can connect
The newly-increased task transformation received, it ensure that change of the system to system load is made and timely respond, avoid data center
Two of Task Congestion and idle running are extreme.
Preferably, the equivalent increment ZL of power seriesu,vComputational methods be：
Using above scheme, the equivalent increment ZL of power seriesu,vValue it is more accurate, it is dynamic so as to substantially increase the present invention
The accuracy of state control.
Preferably, judge whether be excessive value token variable YCu,vObtaining value method be：
Wherein a is given parameters, 1<a<2；Max { } is that maximum operation is asked in set；|ZLu,v| represent ZLu,vAbsolute value；
Avg values are：Avg=mean | ZLu,v||0<u<v≤x}。
Using above scheme, the token variable YC of excessive valueu,vValue it is more accurate, so as to substantially increase the present invention
The accuracy of dynamic control.
Further, after waiting setting time t, 100 milliseconds<t<1000 milliseconds, return and perform step 1, until cloud data
Center is out of service.
Another technical problem to be solved by the present invention is that provide a kind of optimization cloud data center task management performance and calculating
The newly-increased number of tasks device for controlling dynamically of the efficiency of resource management.
To achieve the above object, the invention provides a kind of cloud data center to increase number of tasks device for controlling dynamically newly, including
Data obtaining module, analysis decision module and task management module；The analysis decision module is used for the system information to acquisition
Analyzed and decision-making；The analysis decision module is by data pre-processing unit, controlled quentity controlled variable analytic unit and control decision unit
Composition；First output end of described information acquisition module connects the input of the data pre-processing unit, and described information obtains
Second output end of module connects the first input end of the controlled quentity controlled variable analytic unit, the 3rd output of described information acquisition module
End connects the first input end of the control decision unit, and the output end of the data pre-processing unit connects the controlled quentity controlled variable point
The second input of unit is analysed, the output end of the controlled quentity controlled variable analytic unit connects the second input of the control decision unit
End, the output end of the control decision unit connect the input of the task management module；
Described information acquisition module is used to obtain system information and is sent to the data pre-processing unit, controlled quentity controlled variable analysis
Unit and control decision unit；The system information includes：The newly-increased task quantity XZT of epicycle data centerx, in epicycle data
The task quantity WCT that the heart is completedx, in data center the task that each physical machine is currently run quantity D QTiWith each physical machine
In the task immigration number QYT that epicycle occursi；Wherein 1≤i≤WLJ, WLJ are the quantity of physical machine in data center；
The data pre-processing unit is used to carry out data prediction to the system information got：
Calculate the tasks carrying rate ZXL of data center's epicyclex, calculate each physical machine and can receive new task and account for data center
The ratio BL of overall newly-increased number of tasksi, calculate ZXL1To ZXLxThe step-length average increment INC for eliminating excessive value interference of sequence,
Calculate expected future cloud data center tasks carrying rate FZXL, calculate the expected future execution of each physical machine in cloud data center
Rate FZXLPMi；
The data that pretreatment obtains are sent to the controlled quentity controlled variable analytic unit by the data pre-processing unit；
The controlled quentity controlled variable analytic unit is used to calculate control decision reference value according to the data received：
CalculateObtain each physical machine future it is pre-
Phase remaining acceptable number of tasks FZXLPM；The ZDT is the supported maximum task quantity of single physical machine；
Cloud data center next round is calculated to be expected newly to enter number of tasks KXZT：
The control decision reference value being calculated is sent to the control decision unit by the controlled quentity controlled variable analytic unit；
The control decision unit is used for according to the control decision reference value generation decision information received：
Calculate cloud data center next stage newly-increased task window controlled quentity controlled variable KZL：
The control decision unit increases the cloud data center next stage of generation newly task window controlled quentity controlled variable KZL and is sent to
The task management module；
The task management module is used to increase task management newly：The task management module is according to the decision information received
Judge whether KZL is 0；If KZL is 0, in next round, cloud data center refuses all newly-increased tasks；If KZL is more than 0,
Next round, cloud data center receives first to KZL newly-increased task, and refuses all newly-increased tasks after KZL.
Preferably, the data pre-processing unit passes through calculating
Obtain the task of data center's epicycle
Implementation rate ZXLx；
Mean { } is the operation that set is averaging, and x is epicycle number, 1≤x<Between ∞, t are between often wheel set in advance
Every the time；
The data pre-processing unit passes through calculatingObtaining each physical machine can receive newly
Task accounts for the ratio BL that data center totally increases number of tasks newlyi；
The data pre-processing unit passes through calculating
Obtain eliminating excessive value interference
ZXL afterwards1To ZXLxThe step-length average increment INC of sequence；
The YCu,vFor judge whether be excessive value token variable, ZLu,vRepresent in sequence between u and v record value
The equivalent increment of power series of formation；U and v is positive integer；
The data pre-processing unit passes through calculating
Obtain the equivalent increment ZL of the power seriesu,v；
The data pre-processing unit passes through calculating
Obtain token variable YCu,v；
Wherein a is given parameters, 1<a<2；Max { } is that maximum operation is asked in set；|ZLu,v| represent ZLu,vAbsolute value；
Avg values are：Avg=mean | ZLu,v||0<u<v≤x}；
The data pre-processing unit passes through calculatingObtain expected future cloud
Data center tasks carrying rate FZXL；
The γ is the distance reduction factor, and its effect is nearest historical record value is had a great influence FZXL, and is got over
Early influence is smaller；γ meets 0.5<γ<1；
The data pre-processing unit is by calculating FZXLPMi=FZXL × BLiObtain each physical machine in cloud data center
Expected future implementation rate FZXLPMi。
The present invention also technical problems to be solved, which are to provide a kind of optimization cloud data center task management performance and calculated, to be provided
The newly-increased number of tasks kinetic-control system of the efficiency of source control.
To achieve the above object, the invention provides a kind of cloud data center to increase number of tasks kinetic-control system newly, including
The server of IaaS cloud system, cloud data center is provided with the server and increases number of tasks device for controlling dynamically, the cloud newly
Data center, which increases number of tasks device for controlling dynamically newly, includes data obtaining module, analysis decision module and task management module；Institute
Analysis decision module is stated to be used to analyze the system information of acquisition and decision-making；The analysis decision module is by data prediction
Unit, controlled quentity controlled variable analytic unit and control decision unit composition；First output end of described information acquisition module connects the number
The input of Data preprocess unit, the second output end of described information acquisition module connect the first of the controlled quentity controlled variable analytic unit
Input, the 3rd output end of described information acquisition module connect the first input end of the control decision unit, the data
The output end of pretreatment unit connects the second input of the controlled quentity controlled variable analytic unit, the output of the controlled quentity controlled variable analytic unit
End connects the second input of the control decision unit, and the output end of the control decision unit connects the task management mould
The input of block；
Described information acquisition module is used to obtain system information and is sent to the data pre-processing unit, controlled quentity controlled variable analysis
Unit and control decision unit；The system information includes：The newly-increased task quantity XZT of epicycle data centerx, in epicycle data
The task quantity WCT that the heart is completedx, in data center the task that each physical machine is currently run quantity D QTiWith each physical machine
In the task immigration number QYT that epicycle occursi；Wherein 1≤i≤WLJ, WLJ are the quantity of physical machine in data center；
The data pre-processing unit is used to carry out data prediction to the system information got：
Calculate the tasks carrying rate ZXL of data center's epicyclex, calculate each physical machine and can receive new task and account for data center
The ratio BL of overall newly-increased number of tasksi, calculate ZXL1To ZXLxThe step-length average increment INC for eliminating excessive value interference of sequence,
Calculate expected future cloud data center tasks carrying rate FZXL, calculate the expected future execution of each physical machine in cloud data center
Rate FZXLPMi；
The data that pretreatment obtains are sent to the controlled quentity controlled variable analytic unit by the data pre-processing unit；
The controlled quentity controlled variable analytic unit is used to calculate control decision reference value according to the data received：
CalculateObtain each physical machine future it is pre-
Phase remaining acceptable number of tasks FZXLPM；The ZDT is the supported maximum task quantity of single physical machine；
Cloud data center next round is calculated to be expected newly to enter number of tasks KXZT：
The control decision reference value being calculated is sent to the control decision unit by the controlled quentity controlled variable analytic unit；
The control decision unit is used for according to the control decision reference value generation decision information received：
Calculate cloud data center next stage newly-increased task window controlled quentity controlled variable KZL：
The control decision unit increases the cloud data center next stage of generation newly task window controlled quentity controlled variable KZL and is sent to
The task management module；
The task management module is used to increase task management newly：The task management module is according to the decision information received
Judge whether KZL is 0；If KZL is 0, in next round, cloud data center refuses all newly-increased tasks；If KZL is more than 0,
Next round, cloud data center receives first to KZL newly-increased task, and refuses all newly-increased tasks after KZL.
Preferably, the data pre-processing unit passes through calculating
Obtain the task of data center's epicycle
Implementation rate ZXLx；
Mean { } is the operation that set is averaging, and x is epicycle number, 1≤x<Between ∞, t are between often wheel set in advance
Every the time；
The data pre-processing unit passes through calculatingObtaining each physical machine can receive newly
Task accounts for the ratio BL that data center totally increases number of tasks newlyi；
The data pre-processing unit passes through calculating
Obtain eliminating excessive value interference
ZXL afterwards1To ZXLxThe step-length average increment INC of sequence；
The YCu,vFor judge whether be excessive value token variable, ZLu,vRepresent in sequence between u and v record value
The equivalent increment of power series of formation；U and v is positive integer；
The data pre-processing unit passes through calculating
Obtain the equivalent increment ZL of the power seriesu,v；
The data pre-processing unit passes through calculating
Obtain token variable YCu,v；
Wherein a is given parameters, 1<a<2；Max { } is that maximum operation is asked in set；|ZLu,v| represent ZLu,vAbsolute value；
Avg values are：Avg=mean | ZLu,v||0<u<v≤x}；
The data pre-processing unit passes through calculatingObtain expected future cloud
Data center tasks carrying rate FZXL；
The γ is the distance reduction factor, and its effect is nearest historical record value is had a great influence FZXL, and is got over
Early influence is smaller；γ meets 0.5<γ<1；
The data pre-processing unit is by calculating FZXLPMi=FZXL × BLiObtain each physical machine in cloud data center
Expected future implementation rate FZXLPMi。
The beneficial effects of the invention are as follows：Load and physical machine resource when the present invention is by tracking cloud data center actual motion
Variation tendency, it is dynamic to determine that new task window is big then according to the predicted value that newly-increased task amount can be born to system in future
It is small, avoid Task Congestion and computing resource from dallying, reach optimization balance, taken into account the performance and economy of cloud system.
Brief description of the drawings
Fig. 1 is the flow signal that cloud data center of the present invention increases the embodiment of number of tasks dynamic control method one newly
Figure.
Fig. 2 is that the circuit theory of the newly-increased embodiment of number of tasks device for controlling dynamically one of cloud data center of the present invention is shown
It is intended to.
Fig. 3 is that the circuit theory of the newly-increased embodiment of number of tasks kinetic-control system one of cloud data center of the present invention is shown
It is intended to.
Fig. 4 is the performance comparision figure that cloud data center increases number of tasks dynamic control method newly.
Embodiment
The invention will be further described with reference to the accompanying drawings and examples：
As shown in figure 1, a kind of cloud data center increases number of tasks dynamic control method newly, according to interval time t to cloud data
Center carries out task distribution, carries out according to the following steps：
Step 1: obtaining system information, the system information is cloud data center information.The system information includes：This
Take turns the newly-increased task quantity XZT of data centerx, epicycle data center complete task quantity WCTx, each physics in data center
The quantity D QT for the task that machine is currently runiThe task immigration number QYT occurred with each physical machine in epicyclei.Wherein 1≤i≤
WLJ, WLJ are the quantity of physical machine in data center.Wherein subscript x represents wheel number, due to according to Fixed Time Interval system
System analysis and control, the operation of each secondary control is considered as a wheel, after system is since the first round, as long as not being stopped, takes turns number
Just it is continuously increased.
Step 2: being analyzed the system information of acquisition and decision-making.
The system information of described pair of acquisition is analyzed and decision-making is carried out according to the following steps：
A1, data prediction is carried out to the system information got.
Calculate the tasks carrying rate ZXL of data center's epicyclex, calculate each physical machine and can receive new task and account for data center
The ratio BL of overall newly-increased number of tasksi, calculate ZXL1To ZXLxThe step-length average increment INC for eliminating excessive value interference of sequence,
Calculate expected future cloud data center tasks carrying rate FZXL, calculate the expected future execution of each physical machine in cloud data center
Rate FZXLPMi。
Calculate the tasks carrying rate ZXL of data center's epicyclex：
Mean { } is the operation that set is averaging, and x is epicycle number, 1≤x<Between ∞, t are between often wheel set in advance
Every the time.The meaning directly perceived of above-mentioned formula is, if the task quantity completed of epicycle plus all physical machines migration task quantity it
It is not 0, then is gone through in the past using the tasks carrying rate that above-mentioned quantity divided by interval time t are epicycle data center, on the contrary then basis
The average task handling rate of Records of the Historian record is the implementation rate of epicycle.
The ratio BL that new task accounts for the newly-increased number of tasks of data center's totality can be received by calculating each physical machinei：
The meaning directly perceived of above-mentioned formula is that each physical machine can receive new task and account for number
Totally increase the ratio of number of tasks newly according to center, the physics can be calculated as in the remaining acceptable number of tasks of epicycle divided by all physics
The ratio of machine residue number of tasks sum.
Calculate ZXL1To ZXLxThe step-length average increment INC for eliminating excessive value interference of sequence：
The YCu,vFor judge whether be excessive value token variable, ZLu,vRepresent in sequence between u and v record value
The equivalent increment of power series of formation.U and v is positive integer.The meaning directly perceived of above-mentioned formula is：If all two neighboring power series
The increment formed between equivalent increment is all excessive value, then average be used as choosing non-adjacent value increment eliminates excessive value interference
Step-length average increment.Conversely, the step-length average increment for eliminating excessive value interference is then used as using the average of consecutive value increment.Institute
Stating excessive value interference includes excessive value interference and the interference of too small value.
Next expected future cloud data center tasks carrying rate FZXL is calculated：
The γ is the distance reduction factor, and its effect is nearest historical record value is had a great influence FZXL, and is got over
Early influence is smaller.γ meets 0.5<γ<1.γ values are 0.9 in the present embodiment.
Calculate the expected future implementation rate FZXLPM of each physical machine in cloud data centeri：
FZXLPMi=FZXL × BLi。
A2, calculate control decision reference value.
CalculateEach physical machine will be obtained in future
It is expected that remaining acceptable number of tasks FZXLPM.The ZDT is the supported maximum task quantity of single physical machine.Above-mentioned formula
Intuitively meaning is：The expected remaining acceptable number of tasks of i-th of physical machine next round, is calculated as single physical machine is supported
Maximum task quantity, subtract the task quantity DQT run in i-th of physical machine of epicyclei, then the next round physical machine is subtracted from cloud
The mathematical expectation for the newly-increased task that data center is assigned to.
Cloud data center next round is calculated to be expected newly to enter number of tasks KXZT：
A3, send decision information.
Calculate cloud data center next stage newly-increased task window controlled quentity controlled variable KZL：
The meaning directly perceived of above-mentioned formula is that next stage increases task window controlled quentity controlled variable newly and can be calculated as：When next round is expected
Newly enter number of tasks KXZT and increase number of tasks newly plus epicycle data center, then subtract epicycle data center task and complete number, as a result
During less than zero, window control amount is arranged to 0.When above-mentioned result of calculation is more than 0, but number of tasks can be supported less than data center's maximum
When summation is multiplied by factor beta, then counts epicycle and be not up to the maximum physical machine quantity for supporting the number of tasks upper limit of single physical machine (i.e.), window control amount is then set to this value.When first two situation is all unsatisfactory for, window control amount is set
For " the next round stage is expected newly to enter number of tasks KXZT plus the newly-increased number of tasks of epicycle data center, then subtracts in epicycle data
Heart task completes the result of number ".
Step 3: newly-increased task management：If KZL is 0, in next round, cloud data center refuses all newly-increased tasks.If
KZL is more than 0, then in next round, cloud data center receives first to KZL newly-increased task, and refuses the institute after KZL
There is newly-increased task.
Step 4: after waiting setting time t, 100 milliseconds<t<1000 milliseconds, return and perform step 1, until in cloud data
The heart is out of service.
The equivalent increment ZL of power seriesu,vComputational methods be：
Judge whether be excessive value token variable YCu,vObtaining value method be：
Wherein a is given parameters, 1<a<2, in the present embodiment, a 1.5.Max { } is that maximum operation is asked in set.|ZLu,v
| represent ZLu,vAbsolute value.Avg values are：Avg=mean | ZLu,v||0<u<v≤x}。
As shown in Fig. 2 a kind of cloud data center increases number of tasks device for controlling dynamically, including data obtaining module 3, analysis newly
Decision-making module 4 and task management module 5.The analysis decision module 4 is used to analyze the system information of acquisition and decision-making.
The analysis decision module 4 is made up of data pre-processing unit 401, controlled quentity controlled variable analytic unit 402 and control decision unit 403.
First output end of described information acquisition module 3 connects the input of the data pre-processing unit 401, and described information obtains mould
The first input end of the second output end connection controlled quentity controlled variable analytic unit 402 of block 3, the 3rd of described information acquisition module 3 the
Output end connects the first input end of the control decision unit 403, the output end connection institute of the data pre-processing unit 401
The second input of controlled quentity controlled variable analytic unit 402 is stated, the output end of the controlled quentity controlled variable analytic unit 402 connects the control decision
Second input of unit 403, the output end of the control decision unit 403 connect the input of the task management module 5.
Described information acquisition module 3 is used to obtain system information and is sent to the data pre-processing unit 401, controlled quentity controlled variable
Analytic unit 402 and control decision unit 403.The system information includes：The newly-increased task quantity XZT of epicycle data centerx、
The task quantity WCT that epicycle data center completesx, in data center the task that each physical machine is currently run quantity D QTiWith
The task immigration number QYT that each physical machine occurs in epicyclei.Wherein 1≤i≤WLJ, WLJ are physical machine in data center
Quantity.
The data pre-processing unit 401 is used to carry out data prediction to the system information got：
Calculate the tasks carrying rate ZXL of data center's epicyclex, calculate each physical machine and can receive new task and account for data center
The ratio BL of overall newly-increased number of tasksi, calculate ZXL1To ZXLxThe step-length average increment INC for eliminating excessive value interference of sequence,
Calculate expected future cloud data center tasks carrying rate FZXL, calculate the expected future execution of each physical machine in cloud data center
Rate FZXLPMi。
The data that pretreatment obtains are sent to the controlled quentity controlled variable analytic unit 402 by the data pre-processing unit 401.
The controlled quentity controlled variable analytic unit 402 is used to calculate control decision reference value according to the data received：
CalculateObtain each physical machine future it is pre-
Phase remaining acceptable number of tasks FZXLPM.The ZDT is the supported maximum task quantity of single physical machine.
Cloud data center next round is calculated to be expected newly to enter number of tasks KXZT：
The control decision reference value being calculated is sent to the control decision unit by the controlled quentity controlled variable analytic unit 402
403。
The control decision unit 403 is used for according to the control decision reference value generation decision information received：
Calculate cloud data center next stage newly-increased task window controlled quentity controlled variable KZL：
The control decision unit 403 increases the cloud data center next stage of generation newly task window controlled quentity controlled variable KZL hairs
Give the task management module 5.
The task management module 5 is used to increase task management newly：The task management module 5 is believed according to the decision-making received
Breath judges whether KZL is 0.If KZL is 0, in next round, cloud data center refuses all newly-increased tasks.If KZL is more than 0,
In next round, cloud data center receives first to KZL newly-increased task, and refuses all newly-increased tasks after KZL.
The data pre-processing unit 401 passes through calculating
Obtain the task of data center's epicycle
Implementation rate ZXLx。
Mean { } is the operation that set is averaging, and x is epicycle number, 1≤x<Between ∞, t are between often wheel set in advance
Every the time.
The data pre-processing unit 401 passes through calculatingObtaining each physical machine can receive
New task accounts for the ratio BL that data center totally increases number of tasks newlyi。
The data pre-processing unit 401 passes through calculating
Obtain eliminating excessive value interference
ZXL afterwards1To ZXLxThe step-length average increment INC of sequence.
The YCu,vFor judge whether be excessive value token variable, ZLu,vRepresent in sequence between u and v record value
The equivalent increment of power series of formation.U and v is positive integer.
The data pre-processing unit 401 passes through calculating
Obtain the equivalent increment ZL of the power seriesu,v。
The data pre-processing unit 401 passes through calculating
Obtain token variable YCu,v。
Wherein a is given parameters, 1<a<2.Max { } is that maximum operation is asked in set.|ZLu,v| represent ZLu,vAbsolute value.
Avg values are：Avg=mean | ZLu,v||0<u<v≤x}。
The data pre-processing unit 401 passes through calculatingObtain following pre-
Phase cloud data center tasks carrying rate FZXL.
The γ is the distance reduction factor, and its effect is nearest historical record value is had a great influence FZXL, and is got over
Early influence is smaller.γ meets 0.5<γ<1.γ values are 0.9 in the present embodiment.
The data pre-processing unit 401 is by calculating FZXLPMi=FZXL × BLiObtain each thing in cloud data center
The expected future implementation rate FZXLPM of reason machinei。
As shown in figure 3, cloud data center provided by the invention increases number of tasks device for controlling dynamically newly, one can be deployed in
In existing server, the cloud data center that is exclusively used in being separately provided with one can also be disposed and increase number of tasks dynamic control newly
In server.Therefore, the invention provides a kind of cloud data center to increase number of tasks kinetic-control system, including IaaS cloud system newly
Server 1, cloud data center is provided with the server 1 and increases number of tasks device for controlling dynamically 2, the cloud data center newly
Newly-increased number of tasks device for controlling dynamically 2 includes data obtaining module 3, analysis decision module 4 and task management module 5.Described point
Analysis decision-making module 4 is used to analyze the system information of acquisition and decision-making.The analysis decision module 4 is by data prediction list
Member 401, controlled quentity controlled variable analytic unit 402 and control decision unit 403 form.First output end of described information acquisition module 3 connects
The input of the data pre-processing unit 401 is connect, the second output end of described information acquisition module 3 connects the controlled quentity controlled variable point
The first input end of unit 402 is analysed, the 3rd output end of described information acquisition module 3 connects the control decision unit 403
First input end, the output end of the data pre-processing unit 401 connect the second input of the controlled quentity controlled variable analytic unit 402
End, the output end of the controlled quentity controlled variable analytic unit 402 connect the second input of the control decision unit 403, the control
The output end of decision package 403 connects the input of the task management module 5.
Described information acquisition module 3 is used to obtain system information and is sent to the data pre-processing unit 401, controlled quentity controlled variable
Analytic unit 402 and control decision unit 403.The system information includes：The newly-increased task quantity XZT of epicycle data centerx、
The task quantity WCT that epicycle data center completesx, in data center the task that each physical machine is currently run quantity D QTiWith
The task immigration number QYT that each physical machine occurs in epicyclei.Wherein 1≤i≤WLJ, WLJ are physical machine in data center
Quantity.
The data pre-processing unit 401 is used to carry out data prediction to the system information got：
Calculate the tasks carrying rate ZXL of data center's epicyclex, calculate each physical machine and can receive new task and account for data center
The ratio BL of overall newly-increased number of tasksi, calculate ZXL1To ZXLxThe step-length average increment INC for eliminating excessive value interference of sequence,
Calculate expected future cloud data center tasks carrying rate FZXL, calculate the expected future execution of each physical machine in cloud data center
Rate FZXLPMi。
The data that pretreatment obtains are sent to the controlled quentity controlled variable analytic unit 402 by the data pre-processing unit 401.
The controlled quentity controlled variable analytic unit 402 is used to calculate control decision reference value according to the data received：
CalculateEach physical machine will be obtained in future
It is expected that remaining acceptable number of tasks FZXLPM.The ZDT is the supported maximum task quantity of single physical machine.
Cloud data center next round is calculated to be expected newly to enter number of tasks KXZT：
The control decision reference value being calculated is sent to the control decision unit by the controlled quentity controlled variable analytic unit 402
403。
The control decision unit 403 is used for according to the control decision reference value generation decision information received：
Calculate cloud data center next stage newly-increased task window controlled quentity controlled variable KZL：
The control decision unit 403 increases the cloud data center next stage of generation newly task window controlled quentity controlled variable KZL hairs
Give the task management module 5.
The task management module 5 is used to increase task management newly：The task management module 5 is believed according to the decision-making received
Breath judges whether KZL is 0.If KZL is 0, in next round, cloud data center refuses all newly-increased tasks.If KZL is more than 0,
In next round, cloud data center receives first to KZL newly-increased task, and refuses all newly-increased tasks after KZL.
The data pre-processing unit 401 passes through calculating
Obtain the task of data center's epicycle
Implementation rate ZXLx。
Mean { } is the operation that set is averaging, and x is epicycle number, 1≤x<Between ∞, t are between often wheel set in advance
Every the time.
The data pre-processing unit 401 passes through calculatingObtaining each physical machine can receive
New task accounts for the ratio BL that data center totally increases number of tasks newlyi。
The data pre-processing unit 401 passes through calculating
Obtain eliminating excessive value interference
ZXL afterwards1To ZXLxThe step-length average increment INC of sequence.
The YCu,vFor judge whether be excessive value token variable, ZLu,vRepresent in sequence between u and v record value
The equivalent increment of power series of formation.U and v is positive integer.
The data pre-processing unit 401 passes through calculating
Obtain the equivalent increment ZL of the power seriesu,v。
The data pre-processing unit 401 passes through calculating
Obtain token variable YCu,v。
Wherein a is given parameters, 1<a<2.Max { } is that maximum operation is asked in set.|ZLu,v| represent ZLu,vAbsolute value.
Avg values are：Avg=mean | ZLu,v||0<u<v≤x}。
The data pre-processing unit 401 passes through calculatingObtain following pre-
Phase cloud data center tasks carrying rate FZXL.
The γ is the distance reduction factor, and its effect is nearest historical record value is had a great influence FZXL, and is got over
Early influence is smaller.γ meets 0.5<γ<1.In the present embodiment, γ values are 0.9.
The data pre-processing unit 401 is by calculating FZXLPMi=FZXL × BLiObtain each thing in cloud data center
The expected future implementation rate FZXLPM of reason machinei。
The present invention has advantages below relative to traditional cloud system task load control method：Realize the cloud of flexibility
Increase task window and related management strategy in data newly, loaded according to the real time execution of each physical machine, dynamically determined down
One stage admissible newly-increased task transformation.It can guarantee that change of the system to system load is made timely to respond, make number
Two for avoiding Task Congestion according to center and dallying are extreme, it is ensured that performance and the balance of energy saving economy.
In order to verify above-mentioned cloud data center increase newly number of tasks dynamic control method, device and system actual effect, I
Built a small-scale cloud data center system, system using 4 work stations (concrete configuration as：DELL T3610 work stations,
To strong E5-1630 processors, 8G internal memories, 500GB hard disks) it is the cloud system that core builds 4 physical machines.The processing of each physical machine
Ability is set as：It can support to could support up 50 virtual machine processes, each virtual machine process can at most take 64M internal memories and 200M
Hard-disc storage, it is 10 minutes that the business of each most long permission of virtual machine process, which performs the time,.One is run in the data center
The individual student towards University Of Chongqing undergraduate looks into the business such as class, curricula-variable, coursework are submitted, Course Exercise task is downloaded
System.We have recorded the data center and are using and do not using the cloud data center of the present embodiment to increase number of tasks kinetic controlling equation newly
Performance difference in the case of two kinds of method.
Fig. 4 have recorded the performance ratio taken and do not take the present embodiment cloud data center to increase number of tasks dynamic control method newly
Compared with figure, dotted line mark curve is the curricula-variable task for taking the present embodiment cloud data center to increase number of tasks dynamic control method newly in figure
Average response time test value, solid line are the average response time test value do not taken.Abscissa is to be brought into operation from system
Rise, at the time of obtaining average response time test value.It can be seen that under 30 big course classified resource facilities, dotted line is shown
Task average response time except being slightly longer than solid line when system just brings into operation, its average response time is below reality afterwards
The respective value of line, and difference gradually widens, and embodies obvious performance advantage.
Preferred embodiment of the invention described in detail above.It should be appreciated that one of ordinary skill in the art without
Creative work can is needed to make many modifications and variations according to the design of the present invention.Therefore, all technologies in the art
Personnel are available by logical analysis, reasoning, or a limited experiment on the basis of existing technology under this invention's idea
Technical scheme, all should be in the protection domain being defined in the patent claims.
Claims (8)
1. a kind of cloud data center increases number of tasks dynamic control method newly, task is carried out to cloud data center according to interval time t
Distribution, it is characterised in that carry out according to the following steps：
Step 1: obtain system information；The system information includes：The newly-increased task quantity XZT of epicycle data centerx, epicycle number
The task quantity WCT completed according to centerx, in data center the task that each physical machine is currently run quantity D QTiWith each thing
The task immigration number QYT that reason machine occurs in epicyclei；Wherein 1≤i≤WLJ, WLJ are the quantity of physical machine in data center；
Step 2: being analyzed the system information of acquisition and decision-making；
The system information of described pair of acquisition is analyzed and decision-making is carried out according to the following steps：
A1, data prediction is carried out to the system information got；
Calculate the tasks carrying rate ZXL of data center's epicyclex, calculate each physical machine and can receive new task and account for data center's totality
The ratio BL of newly-increased number of tasksi, calculate ZXL1To ZXLxThe step-length average increment INC for eliminating excessive value interference of sequence, calculate
Expected future cloud data center tasks carrying rate FZXL, the expected future implementation rate for calculating each physical machine in cloud data center
FZXLPMi；
Calculate the tasks carrying rate ZXL of data center's epicyclex：
<mrow>
<msub>
<mi>ZXL</mi>
<mi>x</mi>
</msub>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mfrac>
<mrow>
<msub>
<mi>WCT</mi>
<mi>x</mi>
</msub>
<mo>+</mo>
<munderover>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mi>W</mi>
<mi>L</mi>
<mi>J</mi>
</mrow>
</munderover>
<msub>
<mi>QYT</mi>
<mi>i</mi>
</msub>
</mrow>
<mi>t</mi>
</mfrac>
</mtd>
<mtd>
<mrow>
<mi>i</mi>
<mi>f</mi>
<mi> </mi>
<msub>
<mi>WCT</mi>
<mi>x</mi>
</msub>
<mo>+</mo>
<munderover>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mi>W</mi>
<mi>L</mi>
<mi>J</mi>
</mrow>
</munderover>
<msub>
<mi>QYT</mi>
<mi>i</mi>
</msub>
<mo>></mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mfrac>
<mrow>
<mi>m</mi>
<mi>e</mi>
<mi>a</mi>
<mi>n</mi>
<mo>{</mo>
<msub>
<mi>WCT</mi>
<mi>y</mi>
</msub>
<mo>|</mo>
<mn>0</mn>
<mo><</mo>
<mi>y</mi>
<mo><</mo>
<mi>x</mi>
<mo>}</mo>
</mrow>
<mi>t</mi>
</mfrac>
</mtd>
<mtd>
<mrow>
<mi>e</mi>
<mi>l</mi>
<mi>s</mi>
<mi>e</mi>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>;</mo>
</mrow>
Mean { } is the operation that set is averaging, and x is epicycle number, 1≤x<When ∞, t are the interval between often wheel set in advance
Between；
The ratio BL that new task accounts for the newly-increased number of tasks of data center's totality can be received by calculating each physical machinei：
<mrow>
<msub>
<mi>BL</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<msub>
<mi>ZDT</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<msub>
<mi>DQT</mi>
<mi>i</mi>
</msub>
</mrow>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mi>W</mi>
<mi>L</mi>
<mi>J</mi>
</mrow>
</munderover>
<mrow>
<mo>(</mo>
<msub>
<mi>ZDT</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<msub>
<mi>DQT</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
</mfrac>
<mo>;</mo>
</mrow>
Calculate ZXL1To ZXLxThe step-length average increment INC for eliminating excessive value interference of sequence：
<mrow>
<mi>I</mi>
<mi>N</mi>
<mi>C</mi>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<mi>m</mi>
<mi>e</mi>
<mi>a</mi>
<mi>n</mi>
<mo>{</mo>
<msub>
<mi>ZL</mi>
<mrow>
<mi>u</mi>
<mo>,</mo>
<mi>u</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>|</mo>
<mn>0</mn>
<mo><</mo>
<mi>u</mi>
<mo>&le;</mo>
<mi>x</mi>
<mo>}</mo>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>i</mi>
<mi>f</mi>
<munder>
<mi>&Pi;</mi>
<mrow>
<mn>0</mn>
<mo><</mo>
<mi>u</mi>
<mo>&le;</mo>
<mi>x</mi>
</mrow>
</munder>
<msub>
<mi>YC</mi>
<mrow>
<mi>u</mi>
<mo>,</mo>
<mi>u</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>=</mo>
<mn>1</mn>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mi>m</mi>
<mi>e</mi>
<mi>a</mi>
<mi>n</mi>
<mo>{</mo>
<msub>
<mi>ZL</mi>
<mrow>
<mi>u</mi>
<mo>,</mo>
<mi>v</mi>
</mrow>
</msub>
<mo>|</mo>
<mn>0</mn>
<mo><</mo>
<mi>u</mi>
<mo><</mo>
<mi>v</mi>
<mo>&le;</mo>
<mi>x</mi>
<mo>,</mo>
<mi>v</mi>
<mo>-</mo>
<mi>u</mi>
<mo>></mo>
<mn>1</mn>
<mo>,</mo>
<msub>
<mi>YC</mi>
<mrow>
<mi>u</mi>
<mo>,</mo>
<mi>v</mi>
</mrow>
</msub>
<mo>=</mo>
<mn>0</mn>
<mo>}</mo>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>e</mi>
<mi>l</mi>
<mi>s</mi>
<mi>e</mi>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>;</mo>
</mrow>
The YCu,vFor judge whether be excessive value token variable, ZLu,vRepresent to be formed between u and v record value in sequence
The equivalent increment of power series；U and v is positive integer；
Next expected future cloud data center tasks carrying rate FZXL is calculated：
<mrow>
<mi>F</mi>
<mi>Z</mi>
<mi>X</mi>
<mi>L</mi>
<mo>=</mo>
<mfrac>
<mrow>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mn>1</mn>
<mo>&le;</mo>
<mi>u</mi>
<mo>&le;</mo>
<mi>x</mi>
</mrow>
</munder>
<mrow>
<mo>(</mo>
<msub>
<mi>ZXL</mi>
<mi>u</mi>
</msub>
<mo>+</mo>
<msup>
<mi>INC</mi>
<mrow>
<mi>x</mi>
<mo>-</mo>
<mi>u</mi>
</mrow>
</msup>
<mo>)</mo>
</mrow>
<mo>&times;</mo>
<msup>
<mi>&gamma;</mi>
<mrow>
<mi>x</mi>
<mo>-</mo>
<mi>u</mi>
</mrow>
</msup>
</mrow>
<mrow>
<munder>
<mo>&Sigma;</mo>
<mrow>
<mn>1</mn>
<mo>&le;</mo>
<mi>u</mi>
<mo>&le;</mo>
<mi>x</mi>
</mrow>
</munder>
<msup>
<mi>&gamma;</mi>
<mrow>
<mi>x</mi>
<mo>-</mo>
<mi>u</mi>
</mrow>
</msup>
</mrow>
</mfrac>
<mo>;</mo>
</mrow>
The γ is the distance reduction factor, and its effect is nearest historical record value is had a great influence FZXL, and more early
Influence smaller；γ meets 0.5<γ<1；
Calculate the expected future implementation rate FZXLPM of each physical machine in cloud data centeri：
FZXLPMi=FZXL × BLi；
A2, calculate control decision reference value；
CalculateThe expection that each physical machine will be obtained in future remains
Remaining acceptable number of tasks FZXLPM；The ZDT is the supported maximum task quantity of single physical machine；
Cloud data center next round is calculated to be expected newly to enter number of tasks KXZT：
<mrow>
<mi>K</mi>
<mi>X</mi>
<mi>Z</mi>
<mi>T</mi>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mi>W</mi>
<mi>L</mi>
<mi>J</mi>
</mrow>
</munderover>
<mi>F</mi>
<mi>X</mi>
<mi>Z</mi>
<mi>L</mi>
<mi>P</mi>
<mi>M</mi>
<mo>;</mo>
</mrow>
A3, send decision information；
Calculate cloud data center next stage newly-increased task window controlled quentity controlled variable KZL：
<mrow>
<mi>K</mi>
<mi>Z</mi>
<mi>L</mi>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mn>1</mn>
</mtd>
<mtd>
<mrow>
<mi>i</mi>
<mi>f</mi>
<mi> </mi>
<mi>K</mi>
<mi>X</mi>
<mi>Z</mi>
<mi>T</mi>
<mo>+</mo>
<msub>
<mi>XZT</mi>
<mi>x</mi>
</msub>
<mo>-</mo>
<msub>
<mi>WCT</mi>
<mi>x</mi>
</msub>
<mo><</mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<munderover>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
<mo>,</mo>
<mi>Z</mi>
<mi>D</mi>
<mi>T</mi>
<mo>-</mo>
<msub>
<mi>DQT</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<msub>
<mi>QYT</mi>
<mi>i</mi>
</msub>
<mo>></mo>
<mn>0</mn>
</mrow>
<mrow>
<mi>W</mi>
<mi>L</mi>
<mi>J</mi>
</mrow>
</munderover>
<mn>1</mn>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>i</mi>
<mi>f</mi>
<mi> </mi>
<mn>0</mn>
<mo>&le;</mo>
<mi>K</mi>
<mi>X</mi>
<mi>Z</mi>
<mi>T</mi>
<mo>+</mo>
<msub>
<mi>XZT</mi>
<mi>x</mi>
</msub>
<mo>-</mo>
<msub>
<mi>WCT</mi>
<mi>x</mi>
</msub>
<mo><</mo>
<mi>W</mi>
<mi>L</mi>
<mi>J</mi>
<mo>&times;</mo>
<mi>Z</mi>
<mi>D</mi>
<mi>T</mi>
<mo>&times;</mo>
<mi>&beta;</mi>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mi>K</mi>
<mi>X</mi>
<mi>Z</mi>
<mi>T</mi>
<mo>+</mo>
<msub>
<mi>XZT</mi>
<mi>x</mi>
</msub>
<mo>-</mo>
<msub>
<mi>WCT</mi>
<mi>x</mi>
</msub>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>e</mi>
<mi>l</mi>
<mi>s</mi>
<mi>e</mi>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>;</mo>
</mrow>
Step 3: newly-increased task management：If KZL is 0, in next round, cloud data center refuses all newly-increased tasks；If KZL
More than 0, then in next round, cloud data center receives first to KZL newly-increased task, and refuses all after KZL
Newly-increased task.
2. cloud data center as claimed in claim 1 increases number of tasks dynamic control method newly, it is characterized in that：Described power series etc.
Imitate increment ZLu,vComputational methods be：
<mrow>
<msub>
<mi>ZL</mi>
<mrow>
<mi>u</mi>
<mo>,</mo>
<mi>v</mi>
</mrow>
</msub>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mroot>
<mrow>
<msub>
<mi>ZXL</mi>
<mi>v</mi>
</msub>
<mo>-</mo>
<msub>
<mi>ZXL</mi>
<mi>u</mi>
</msub>
</mrow>
<mrow>
<mi>v</mi>
<mo>-</mo>
<mi>u</mi>
</mrow>
</mroot>
</mtd>
<mtd>
<mrow>
<mi>i</mi>
<mi>f</mi>
<mi> </mi>
<msub>
<mi>ZXL</mi>
<mi>v</mi>
</msub>
<mo>-</mo>
<msub>
<mi>ZXL</mi>
<mi>u</mi>
</msub>
<mo>></mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mo>-</mo>
<mroot>
<mrow>
<mo>-</mo>
<mrow>
<mo>(</mo>
<msub>
<mi>ZXL</mi>
<mi>v</mi>
</msub>
<mo>-</mo>
<msub>
<mi>ZXL</mi>
<mi>u</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>v</mi>
<mo>-</mo>
<mi>u</mi>
</mrow>
</mroot>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>e</mi>
<mi>l</mi>
<mi>s</mi>
<mi>e</mi>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>.</mo>
</mrow>
3. cloud data center as claimed in claim 1 increases number of tasks dynamic control method newly, it is characterized in that：Judge whether be
The token variable YC of valueu,vObtaining value method be：
<mrow>
<msub>
<mi>YC</mi>
<mrow>
<mi>u</mi>
<mo>,</mo>
<mi>v</mi>
</mrow>
</msub>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mn>1</mn>
</mtd>
<mtd>
<mrow>
<mi>i</mi>
<mi>f</mi>
<mfrac>
<mrow>
<mo>|</mo>
<msub>
<mi>ZL</mi>
<mrow>
<mi>u</mi>
<mo>,</mo>
<mi>v</mi>
</mrow>
</msub>
<mo>|</mo>
</mrow>
<mrow>
<mi>a</mi>
<mi>v</mi>
<mi>g</mi>
</mrow>
</mfrac>
<mo>></mo>
<mi>a</mi>
<mo>,</mo>
<mo>|</mo>
<msub>
<mi>ZL</mi>
<mrow>
<mi>u</mi>
<mo>,</mo>
<mi>v</mi>
</mrow>
</msub>
<mo>|</mo>
<mo>></mo>
<mi>m</mi>
<mi>a</mi>
<mi>x</mi>
<mo>{</mo>
<msub>
<mi>ZL</mi>
<mrow>
<mi>s</mi>
<mo>,</mo>
<mi>s</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>|</mo>
<mi>u</mi>
<mo>&le;</mo>
<mi>s</mi>
<mo>&le;</mo>
<mi>v</mi>
<mo>}</mo>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mn>0</mn>
</mtd>
<mtd>
<mrow>
<mi>e</mi>
<mi>l</mi>
<mi>s</mi>
<mi>e</mi>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>;</mo>
</mrow>
Wherein a is given parameters, 1<a<2；Max { } is that maximum operation is asked in set；|ZLu,v| represent ZLu,vAbsolute value；avg
Value is：Avg=mean | ZLu,v||0<u<v≤x}。
4. the cloud data center as described in claim 1 or 2 or 3 increases number of tasks dynamic control method newly, it is characterized in that：Wait is set
Fix time after t, 100 milliseconds<t<1000 milliseconds, return and perform step 1, until cloud data center is out of service.
5. a kind of cloud data center increases number of tasks device for controlling dynamically newly, it is characterised in that：Including data obtaining module (3), divide
Analyse decision-making module (4) and task management module (5)；The analysis decision module (4) is used to divide the system information of acquisition
Analysis and decision-making；The analysis decision module (4) is determined by data pre-processing unit (401), controlled quentity controlled variable analytic unit (402) and control
Plan unit (403) forms；First output end of described information acquisition module (3) connects the data pre-processing unit (401)
Input, the second output end of described information acquisition module (3) connect the first input of the controlled quentity controlled variable analytic unit (402)
End, the 3rd output end of described information acquisition module (3) connects the first input end of the control decision unit (403), described
The output end of data pre-processing unit (401) connects the second input of the controlled quentity controlled variable analytic unit (402), the controlled quentity controlled variable
The output end of analytic unit (402) connects the second input of the control decision unit (403), the control decision unit
(403) output end connects the input of the task management module (5)；
Described information acquisition module (3) is used to obtain system information and is sent to the data pre-processing unit (401), controlled quentity controlled variable
Analytic unit (402) and control decision unit (403)；The system information includes：The newly-increased task quantity of epicycle data center
XZTx, epicycle data center complete task quantity WCTx, in data center the task that each physical machine is currently run quantity
DQTiThe task immigration number QYT occurred with each physical machine in epicyclei；Wherein 1≤i≤WLJ, WLJ are physics in data center
The quantity of machine；
The data pre-processing unit (401) is used to carry out data prediction to the system information got：
Calculate the tasks carrying rate ZXL of data center's epicyclex, calculate each physical machine and can receive new task and account for data center's totality
The ratio BL of newly-increased number of tasksi, calculate ZXL1To ZXLxThe step-length average increment INC for eliminating excessive value interference of sequence, calculate
Expected future cloud data center tasks carrying rate FZXL, the expected future implementation rate for calculating each physical machine in cloud data center
FZXLPMi；
The data that pretreatment obtains are sent to the controlled quentity controlled variable analytic unit (402) by the data pre-processing unit (401)；
The controlled quentity controlled variable analytic unit (402) is used to calculate control decision reference value according to the data received：
CalculateThe expection that each physical machine will be obtained in future remains
Remaining acceptable number of tasks FZXLPM；The ZDT is the supported maximum task quantity of single physical machine；
Cloud data center next round is calculated to be expected newly to enter number of tasks KXZT：
<mrow>
<mi>K</mi>
<mi>X</mi>
<mi>Z</mi>
<mi>T</mi>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mi>W</mi>
<mi>L</mi>
<mi>J</mi>
</mrow>
</munderover>
<msub>
<mi>FXZLPM</mi>
<mi>i</mi>
</msub>
<mo>;</mo>
</mrow>
The control decision reference value being calculated is sent to the control decision unit by the controlled quentity controlled variable analytic unit (402)
(403)；
The control decision unit (403) is used for according to the control decision reference value generation decision information received：
Calculate cloud data center next stage newly-increased task window controlled quentity controlled variable KZL：
<mrow>
<mi>K</mi>
<mi>Z</mi>
<mi>L</mi>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mn>1</mn>
</mtd>
<mtd>
<mrow>
<mi>i</mi>
<mi>f</mi>
<mi> </mi>
<mi>K</mi>
<mi>X</mi>
<mi>Z</mi>
<mi>T</mi>
<mo>+</mo>
<msub>
<mi>XZT</mi>
<mi>x</mi>
</msub>
<mo>-</mo>
<msub>
<mi>WCT</mi>
<mi>x</mi>
</msub>
<mo><</mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<munderover>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
<mo>,</mo>
<mi>Z</mi>
<mi>D</mi>
<mi>T</mi>
<mo>-</mo>
<msub>
<mi>DQT</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<msub>
<mi>QYT</mi>
<mi>i</mi>
</msub>
<mo>></mo>
<mn>0</mn>
</mrow>
<mrow>
<mi>W</mi>
<mi>L</mi>
<mi>J</mi>
</mrow>
</munderover>
<mn>1</mn>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>i</mi>
<mi>f</mi>
<mi> </mi>
<mn>0</mn>
<mo>&le;</mo>
<mi>K</mi>
<mi>X</mi>
<mi>Z</mi>
<mi>T</mi>
<mo>+</mo>
<msub>
<mi>XZT</mi>
<mi>x</mi>
</msub>
<mo>-</mo>
<msub>
<mi>WCT</mi>
<mi>x</mi>
</msub>
<mo><</mo>
<mi>W</mi>
<mi>L</mi>
<mi>J</mi>
<mo>&times;</mo>
<mi>Z</mi>
<mi>D</mi>
<mi>T</mi>
<mo>&times;</mo>
<mi>&beta;</mi>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mi>K</mi>
<mi>X</mi>
<mi>Z</mi>
<mi>T</mi>
<mo>+</mo>
<msub>
<mi>XZT</mi>
<mi>x</mi>
</msub>
<mo>-</mo>
<msub>
<mi>WCT</mi>
<mi>x</mi>
</msub>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>e</mi>
<mi>l</mi>
<mi>s</mi>
<mi>e</mi>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>;</mo>
</mrow>
The control decision unit (403) increases the cloud data center next stage of generation newly task window controlled quentity controlled variable KZL and sent
Give the task management module (5)；
The task management module (5) is used to increase task management newly：The task management module (5) is according to the decision-making letter received
Breath judges whether KZL is 0；If KZL is 0, in next round, cloud data center refuses all newly-increased tasks；If KZL is more than 0,
In next round, cloud data center receives first to KZL newly-increased task, and refuses all newly-increased tasks after KZL.
6. a kind of cloud data center as claimed in claim 5 increases number of tasks device for controlling dynamically newly, it is characterised in that：
The data pre-processing unit (401) passes through calculating
Obtain the tasks carrying of data center's epicycle
Rate ZXLx；
Mean { } is the operation that set is averaging, and x is epicycle number, 1≤x<When ∞, t are the interval between often wheel set in advance
Between；
The data pre-processing unit (401) passes through calculatingObtaining each physical machine can receive newly
Task accounts for the ratio BL that data center totally increases number of tasks newlyi；
The data pre-processing unit (401) passes through calculating
Obtain after eliminating excessive value interference
ZXL1To ZXLxThe step-length average increment INC of sequence；
The YCu,vFor judge whether be excessive value token variable, ZLu,vRepresent to be formed between u and v record value in sequence
The equivalent increment of power series；U and v is positive integer；
The data pre-processing unit (401) passes through calculating
Obtain the equivalent increment ZL of the power seriesu,v；
The data pre-processing unit (401) passes through calculating
Obtain token variable YCu,v；
Wherein a is given parameters, 1<a<2；Max { } is that maximum operation is asked in set；|ZLu,v| represent ZLu,vAbsolute value；avg
Value is：Avg=mean | ZLu,v||0<u<v≤x}；
The data pre-processing unit (401) passes through calculatingObtain expected future cloud
Data center tasks carrying rate FZXL；
The γ is the distance reduction factor, and its effect is nearest historical record value is had a great influence FZXL, and more early
Influence smaller；γ meets 0.5<γ<1；
The data pre-processing unit (401) is by calculating FZXLPMi=FZXL × BLiObtain each physics in cloud data center
The expected future implementation rate FZXLPM of machinei。
7. a kind of cloud data center increases number of tasks kinetic-control system newly, include the server (1) of IaaS cloud system, its feature exists
In：Cloud data center is provided with the server (1) and increases number of tasks device for controlling dynamically (2) newly, the cloud data center is new
Increasing number of tasks device for controlling dynamically (2) includes data obtaining module (3), analysis decision module (4) and task management module (5)；
The analysis decision module (4) is used to analyze the system information of acquisition and decision-making；The analysis decision module (4) is by counting
Data preprocess unit (401), controlled quentity controlled variable analytic unit (402) and control decision unit (403) composition；Described information acquisition module
(3) input of the first output end connection data pre-processing unit (401), the second of described information acquisition module (3)
Output end connects the first input end of the controlled quentity controlled variable analytic unit (402), the 3rd output end of described information acquisition module (3)
The first input end of the control decision unit (403) is connected, described in the output end connection of the data pre-processing unit (401)
Second input of controlled quentity controlled variable analytic unit (402), the output end connection control of the controlled quentity controlled variable analytic unit (402) are determined
Second input of plan unit (403), the output end of the control decision unit (403) connect the task management module (5)
Input；
Described information acquisition module (3) is used to obtain system information and is sent to the data pre-processing unit (401), controlled quentity controlled variable
Analytic unit (402) and control decision unit (403)；The system information includes：The newly-increased task quantity of epicycle data center
XZTx, epicycle data center complete task quantity WCTx, in data center the task that each physical machine is currently run quantity
DQTiThe task immigration number QYT occurred with each physical machine in epicyclei；Wherein 1≤i≤WLJ, WLJ are physics in data center
The quantity of machine；
The data pre-processing unit (401) is used to carry out data prediction to the system information got：
Calculate the tasks carrying rate ZXL of data center's epicyclex, calculate each physical machine and can receive new task and account for data center's totality
The ratio BL of newly-increased number of tasksi, calculate ZXL1To ZXLxThe step-length average increment INC for eliminating excessive value interference of sequence, calculate
Expected future cloud data center tasks carrying rate FZXL, the expected future implementation rate for calculating each physical machine in cloud data center
FZXLPMi；
The data that pretreatment obtains are sent to the controlled quentity controlled variable analytic unit (402) by the data pre-processing unit (401)；
The controlled quentity controlled variable analytic unit (402) is used to calculate control decision reference value according to the data received：
CalculateThe expection that each physical machine will be obtained in future remains
Remaining acceptable number of tasks FZXLPM；The ZDT is the supported maximum task quantity of single physical machine；
Cloud data center next round is calculated to be expected newly to enter number of tasks KXZT：
<mrow>
<mi>K</mi>
<mi>X</mi>
<mi>Z</mi>
<mi>T</mi>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mi>W</mi>
<mi>L</mi>
<mi>J</mi>
</mrow>
</munderover>
<msub>
<mi>FXZLPM</mi>
<mi>i</mi>
</msub>
<mo>;</mo>
</mrow>
The control decision reference value being calculated is sent to the control decision unit by the controlled quentity controlled variable analytic unit (402)
(403)；
The control decision unit (403) is used for according to the control decision reference value generation decision information received：
Calculate cloud data center next stage newly-increased task window controlled quentity controlled variable KZL：
<mrow>
<mi>K</mi>
<mi>Z</mi>
<mi>L</mi>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mn>1</mn>
</mtd>
<mtd>
<mrow>
<mi>i</mi>
<mi>f</mi>
<mi> </mi>
<mi>K</mi>
<mi>X</mi>
<mi>Z</mi>
<mi>T</mi>
<mo>+</mo>
<msub>
<mi>XZT</mi>
<mi>x</mi>
</msub>
<mo>-</mo>
<msub>
<mi>WCT</mi>
<mi>x</mi>
</msub>
<mo><</mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<munderover>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
<mo>,</mo>
<mi>Z</mi>
<mi>D</mi>
<mi>T</mi>
<mo>-</mo>
<msub>
<mi>DQT</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<msub>
<mi>QYT</mi>
<mi>i</mi>
</msub>
<mo>></mo>
<mn>0</mn>
</mrow>
<mrow>
<mi>W</mi>
<mi>L</mi>
<mi>J</mi>
</mrow>
</munderover>
<mn>1</mn>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>i</mi>
<mi>f</mi>
<mi> </mi>
<mn>0</mn>
<mo>&le;</mo>
<mi>K</mi>
<mi>X</mi>
<mi>Z</mi>
<mi>T</mi>
<mo>+</mo>
<msub>
<mi>XZT</mi>
<mi>x</mi>
</msub>
<mo>-</mo>
<msub>
<mi>WCT</mi>
<mi>x</mi>
</msub>
<mo><</mo>
<mi>W</mi>
<mi>L</mi>
<mi>J</mi>
<mo>&times;</mo>
<mi>Z</mi>
<mi>D</mi>
<mi>T</mi>
<mo>&times;</mo>
<mi>&beta;</mi>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mi>K</mi>
<mi>X</mi>
<mi>Z</mi>
<mi>T</mi>
<mo>+</mo>
<msub>
<mi>XZT</mi>
<mi>x</mi>
</msub>
<mo>-</mo>
<msub>
<mi>WCT</mi>
<mi>x</mi>
</msub>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>e</mi>
<mi>l</mi>
<mi>s</mi>
<mi>e</mi>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>;</mo>
</mrow>
The control decision unit (403) increases the cloud data center next stage of generation newly task window controlled quentity controlled variable KZL and sent
Give the task management module (5)；
The task management module (5) is used to increase task management newly：The task management module (5) is according to the decision-making letter received
Breath judges whether KZL is 0；If KZL is 0, in next round, cloud data center refuses all newly-increased tasks；If KZL is more than 0,
In next round, cloud data center receives first to KZL newly-increased task, and refuses all newly-increased tasks after KZL.
8. a kind of cloud data center as claimed in claim 7 increases number of tasks kinetic-control system newly, it is characterised in that：The number
Data preprocess unit (401) passes through calculating
Obtain the tasks carrying of data center's epicycle
Rate ZXLx；
Mean { } is the operation that set is averaging, and x is epicycle number, 1≤x<When ∞, t are the interval between often wheel set in advance
Between；
The data pre-processing unit (401) passes through calculatingObtaining each physical machine can receive newly
Task accounts for the ratio BL that data center totally increases number of tasks newlyi；
The data pre-processing unit (401) passes through calculating
Obtain after eliminating excessive value interference
ZXL1To ZXLxThe step-length average increment INC of sequence；
The YCu,vFor judge whether be excessive value token variable, ZLu,vRepresent to be formed between u and v record value in sequence
The equivalent increment of power series；U and v is positive integer；
The data pre-processing unit (401) passes through calculating
Obtain the equivalent increment ZL of the power seriesu,v；
The data pre-processing unit (401) passes through calculating
Obtain token variable YCu,v；
Wherein a is given parameters, 1<a<2；Max { } is that maximum operation is asked in set；|ZLu,v| represent ZLu,vAbsolute value；avg
Value is：Avg=mean | ZLu,v||0<u<v≤x}；
The data pre-processing unit (401) passes through calculatingObtain expected future cloud
Data center tasks carrying rate FZXL；
The γ is the distance reduction factor, and its effect is nearest historical record value is had a great influence FZXL, and more early
Influence smaller；γ meets 0.5<γ<1；
The data pre-processing unit (401) is by calculating FZXLPMi=FZXL × BLiObtain each physics in cloud data center
The expected future implementation rate FZXLPM of machinei。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410315765.4A CN104111875B (en) | 2014-07-03 | 2014-07-03 | Cloud data center increases number of tasks device for controlling dynamically, system and method newly |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410315765.4A CN104111875B (en) | 2014-07-03 | 2014-07-03 | Cloud data center increases number of tasks device for controlling dynamically, system and method newly |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104111875A CN104111875A (en) | 2014-10-22 |
| CN104111875B true CN104111875B (en) | 2017-11-28 |
Family
ID=51708673
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410315765.4A Active CN104111875B (en) | 2014-07-03 | 2014-07-03 | Cloud data center increases number of tasks device for controlling dynamically, system and method newly |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104111875B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104503846B (en) * | 2015-01-22 | 2018-01-30 | 成都派沃特科技有限公司 | A kind of resource management system based on cloud computing system |
| CN105208119B (en) * | 2015-09-21 | 2018-06-22 | 重庆大学 | A kind of cloud data center method for allocating tasks, device and system |
| CN105204961B (en) * | 2015-09-21 | 2018-10-26 | 重庆大学 | Method, device and system for setting check point of cloud data center host |
| CN106201847B (en) * | 2016-06-30 | 2019-04-23 | 重庆大学 | Consider method for allocating tasks, the device and system of the decaying of cloud platform host performance |
| CN109213588A (en) * | 2018-09-17 | 2019-01-15 | 重庆大学 | A kind of cloud data center Batch Arrival task allocation apparatus, system and method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102195886A (en) * | 2011-05-30 | 2011-09-21 | 兰雨晴 | Service scheduling method on cloud platform |
| CN103442087A (en) * | 2013-09-12 | 2013-12-11 | 重庆大学 | Web service system access volume control device and method based on response time trend analysis |
| CN103902379A (en) * | 2012-12-25 | 2014-07-02 | 中国移动通信集团公司 | Task scheduling method and device and server cluster |
-
2014
- 2014-07-03 CN CN201410315765.4A patent/CN104111875B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102195886A (en) * | 2011-05-30 | 2011-09-21 | 兰雨晴 | Service scheduling method on cloud platform |
| CN103902379A (en) * | 2012-12-25 | 2014-07-02 | 中国移动通信集团公司 | Task scheduling method and device and server cluster |
| CN103442087A (en) * | 2013-09-12 | 2013-12-11 | 重庆大学 | Web service system access volume control device and method based on response time trend analysis |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104111875A (en) | 2014-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104111875B (en) | Cloud data center increases number of tasks device for controlling dynamically, system and method newly | |
| CN104636197B (en) | A kind of evaluation method of data center's virtual machine (vm) migration scheduling strategy | |
| CN103999049B (en) | Method and apparatus for predicting virtual machine demand | |
| CN105550323B (en) | Load balance prediction method and prediction analyzer for distributed database | |
| CN104065745A (en) | Cloud computing dynamic resource scheduling system and method | |
| CN103607459B (en) | The dynamic resource monitoring of a kind of cloud computing platform IaaS layer and dispatching method | |
| CN104657220A (en) | Model and method for scheduling for mixed cloud based on deadline and cost constraints | |
| CN103955398B (en) | Virtual machine coexisting scheduling method based on processor performance monitoring | |
| CN110348571A (en) | A kind of neural network model training method, device, chip and system | |
| CN104023042B (en) | Cloud platform resource scheduling method | |
| CN103064744B (en) | The method for optimizing resources that a kind of oriented multilayer Web based on SLA applies | |
| Li et al. | Deep learning based parking prediction on cloud platform | |
| CN103401939A (en) | Load balancing method adopting mixing scheduling strategy | |
| CN103685347B (en) | Method and device for allocating network resources | |
| CN109410054B (en) | Block chain consensus method based on information sharing contribution value in autonomous domain mode | |
| CN104636187A (en) | Virtual machine scheduling method in NUMA (non uniform memory access) architecture and based on load prediction | |
| CN112685153A (en) | Micro-service scheduling method and device and electronic equipment | |
| CN103605578A (en) | Load balance scheduling method based on virtual machine migration | |
| CN104199820A (en) | Cloud platform MapReduce workflow scheduling optimizing method | |
| CN104063282B (en) | Management method, device and server for IaaS cloud variable scale resource pool | |
| CN103345663A (en) | Combinatorial optimization method of electric power system set considering creep speed constraints | |
| CN110994646B (en) | Method, system and storage medium for evaluating running effect of AGC (automatic gain control) adjustment of power grid | |
| CN107506233A (en) | A kind of schedule virtual resources method, apparatus and server | |
| CN108900343A (en) | Local storage-based resource prediction and scheduling method for cloud server | |
| CN108132840A (en) | Resource regulating method and device in a kind of distributed system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB03 | Change of inventor or designer information |
Inventor after: Guo Kunyin Inventor after: Lin Xiqiao Inventor after: Xia Yunni Inventor after: Wang Yuandou Inventor after: Zhu Qingsheng Inventor before: Xia Yunni Inventor before: Luo Xin Inventor before: Zeng Lingqiu Inventor before: Sun Tianhao Inventor before: Zhu Qingsheng |
|
| CB03 | Change of inventor or designer information | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20170928 Address after: 400044 Shapingba District Sha Street, No. 174, Chongqing Applicant after: Chongqing University Applicant after: Electric Power Research Institute of Guangxi Power Grid Ltd Address before: 400045 Shapingba District, Sha Sha Street, No. 174, Chongqing Applicant before: Chongqing University |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200612 Address after: 400060 21 office 30, No.73, Nanping West Road, Nan'an District, Chongqing Co-patentee after: ELECTRIC POWER RESEARCH INSTITUTE, GUANGXI POWER GRID Co.,Ltd. Patentee after: CHONGQING JINYUYUN ENERGY TECHNOLOGY Co.,Ltd. Address before: 400044 Shapingba District Sha Street, No. 174, Chongqing Co-patentee before: ELECTRIC POWER RESEARCH INSTITUTE, GUANGXI POWER GRID Co.,Ltd. Patentee before: Chongqing University |
|
| TR01 | Transfer of patent right |