CN102929687B - A kind of energy-conservation cloud computing data center virtual machine laying method - Google Patents
A kind of energy-conservation cloud computing data center virtual machine laying method Download PDFInfo
- Publication number
- CN102929687B CN102929687B CN201210385961.XA CN201210385961A CN102929687B CN 102929687 B CN102929687 B CN 102929687B CN 201210385961 A CN201210385961 A CN 201210385961A CN 102929687 B CN102929687 B CN 102929687B
- Authority
- CN
- China
- Prior art keywords
- physical server
- virtual machine
- physical
- cpu
- sequence
- 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
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
Energy-conservation cloud computing data center virtual machine laying method of the present invention, comprising: a. sets up physical server region; B. obtain physical server information; C. calculate resources of virtual machine to be created; D. physical server is sorted; E. set up the mapping between virtual machine to be created and physical server; F. judged whether that single physical server satisfies the demands; G. judge whether to exist the physical server sequence meeting the demands in same subregion; H. choose the server sequence of air conditioning energy consumption cost minimization and physical server minimum number. Cloud computing of the present invention data center virtual machine laying method, preferentially selects single physical server to create virtual machine, then selects the physical server in same subregion to create virtual machine; At the above two, all in non-existent situation, the physical server sequence that final choice intensity is the highest creates virtual machine, realizes the energy-conservation placement of cloud computing center virtual machine, and energy-saving effect is remarkable, is convenient to application.
Description
Technical field
The present invention relates to a kind of energy-conservation cloud computing data center virtual machine laying method, in particular, relate in particular toA kind of energy-conservation cloud computing data center virtual machine laying method of to greatest extent all virtual machines to be created being concentrated to placement.
Background technology
Cloud computing (CloudComputing) is grid computing, Distributed Calculation, parallel computation, the network storage, virtualDeng the product of traditional computer and network technical development fusion. It is a kind of business computation model, and it is distributed in calculation taskOn the resource pool that a large amount of physical computers form, make various application systems can obtain as required computing capability, memory space and letterBreath service. Its core is Service Source pond, and it more normally can self and the virtual resources of management, comprises calculatingServer, storage server and bandwidth resources. Cloud computing comprises the service of following level:
Infrastructure serve (InfrastructureasaService is called for short IaaS), refers to that consumer passes throughInternet obtains service from perfect computer based Infrastructure, comprises processing, storage, network and other basic calculating moneySource, user can dispose freely thereon and move any software.
Software serve (SoftwareasaService), and referring to a kind ofly provides software service by InternetPattern, user, without buying software, but rents the software based on Web to provider, carrys out management enterprise business activities.
Platform serves (PlatformasaService is called for short PaaS), refers to research and development of software platform to take as oneBusiness, submits to user with the pattern of SaaS. Paas is the one application of SaaS pattern. The appearance of PaaS can be accelerated sending out of SaaSExhibition, especially accelerates the development rate that SaaS applies.
The cloud data center at present main mode that adopts virtual data center builds, and virtual data center refers to and utilizes clothesBusiness device Intel Virtualization Technology, adopts fictitious host computer independent, isolation mutually that the function that is equal to physical machine is provided, its cost well belowPhysical data center. The core of virtual data center is virtual machine, and so-called virtual machine is exactly computer software, and it runs on thingOn reason hardware or physical computer, it can operation system (being called client operating system) and application program, it have fromOneself virtual hardware. Virtual machine is not emulator and simulator, and they are real computers, can realize and physical computerThe identical function that even exceedes physical computer.
User can use the physical server of Intel Virtualization Technology based on a small amount of that large-scale virtualization services is provided, thereforeGreatly reduce the structure expense of cloud data center. At present, power consumption becomes the main operation cost of cloud computing data center.Taking Google as example, according to existing public information, Google has 36 data centers in the whole world, within 1 year, consumes about 79,000,000,000 thousandWatt-hour electric power; 2011, all-american data center consumed electric energy 10,000 hundred million kilowatt hours altogether, amounts to 7,400,000,000 dollars. Power savingBecome the operation of cloud data center and be badly in need of a great problem solving. In the operation of cloud computing data center, it is huge by itPhysical server cluster for provide for user virtual machine Self-Service, virtual machine be mainly created in each physical server itUpper, the establishment of virtual machine can make physical server cluster produce a large amount of used heat, thereby temperature is around raise, according to measuring and calculating,The highest problem can reach degree more than 50, and the rising of temperature can cause the work of air-conditioning refrigeration system, keeps by air-conditioning refrigeration systemThe normal temperature of data center, the power consumption causing thus, accounts for 40% of cloud data center power consumption. The collection of virtual machineMiddle placement, its core concept is that the establishment of virtual machine is concentrated on less physical server, thereby reduces physics serviceThe generation of device used heat, saves power consumption. Therefore, concentrate and place the power consumption that reduces refrigeration system by efficient virtual machineBecome an energy-conservation important channel of cloud data center.
The virtual machine laying method of current main flow mainly comprises bolus dressing, itemize method, and Load-aware method and internal memory are got close to method.
The basic thought of bolus dressing is to use to the greatest extent less node as target, virtual machine is concentrated on to the joint in part cloud computingOperation on point, in implementation, adopts the maximum priority principles of virtual machine operation number, when being newly-built virtual machine selection placeWhen main frame, select to have the host of maximum quantity virtual machines operation, it is a large amount of virtual that the advantage of bolus dressing is that the method can makeMachine concentrates on the physical node of minority and moves, and can reduce physical server cost, and the deficiency of bolus dressing is that virtual machine too collectsIn, cause resources of virtual machine to seize probability excessive, for ensureing the quality of virtual server, the migration of a large amount of virtual machines and resource are adjustedFull line is essential, can produce so a large amount of expenses.
Itemize method basic thought is that to maximize individual server node available resources be target, and virtual machine is dispersed in to instituteHave on node and move, in implementation, adopt virtual machine operation minimum number priority principle, when being newly-built virtual machine choosingSelecting host is, selects to have the host of minimum number virtual machine operation, and its basic ideas are inspired by cluster load balancing,Be that virtual machine is uniformly distributed by quantity, reduce resources of virtual machine and seize probability, its shortcoming is to place according to virtual machine quantity,Only abstract virtual machine quantity be load value on node, too single, and cannot distinguish different resource (as CPU, inDeposit hard disk etc.) the virtual machine actual loading that node is caused of request, refinement not, is difficult to realize more fine granularity and precisionResource distribution requirements.
The target of Load-aware method is identical with itemize method, makes every effort to maximize the available resources on individual node. Basic DesignThinking is to be subject to the minimum inspiration of node load, and newly-built virtual machine is placed on the node with minimum load and is moved. ImplementationUpper, adopt maximum CPU idleness priority principle, be to select CPU idleness when selecting host for newly-built virtual machineLarge host. Its advantage is identical with itemize method, inspired by load balancing, due to the cpu resource service condition of considering on node,Can reach the cpu resource load balancing within the scope of distributed computing system. Its shortcoming is that the method has only been considered cpu resource, rightThe internal memory of node, the responsible important compositions of clustered node such as network and disk service condition, lack and consider.
The internal memory method of getting close to is that a kind of virtual machine based on the shared perception of internal memory being proposed by doctor TimothyWood is placedSystem, comprises an internal memory recognition system, can effectively judge that one group of internal memory between virtual machine shares potential, and calculate moreEffectively modes of emplacement. In addition, along with load variations, system also will be utilized online migration to optimize virtual machine and place, and its advantage isFinding the identical virtual machine of virtual memory page from cluster wide, is that they move to same clustered node, shared virtual memory,Can improve physical memory utilization rate, save internal memory, the virtual machine that promotes cluster holds quantity. Its shortcoming is because needs are by emptyPlan machine migration realizes virtual machine and places, and procedure is comparatively complicated, and if the virtual machine of shared virtual memory too much, Wu FabaoThe service quality of card virtual server.
For virtual machine laying method set forth above, only itself consider a problem not in conjunction with air-conditioning from virtual machine placementThe factor of refrigeration system energy consumption designs, for bolus dressing, its thought be as far as possible by multiple virtual machines concentrate be placed into lessPhysical server in, thereby realize the minimizing of electric energy, but the method has only been considered physical server concentrating in logic, realIn border, the server of deployment may be positioned at different physics rack regions, and according to said method deployment can make different physical serversThe temperature rise in region, causes the work of refrigeration system, thereby consumes electric energy; Get close to for itemize method, Load-aware, internal memoryMethod, its thought is that virtual machine is distributed to multiple physical nodes, thereby realizes the function of load balancing, these class methods can causeIn extensive scope, the rising of temperature, causes the large area work of refrigeration air-conditioner. Therefore said method cannot in implementation procedureRealize the energy-conservation of air-conditioning refrigeration system.
Summary of the invention
The present invention, in order to overcome the shortcoming of above-mentioned technical problem, has proposed one to greatest extent by all to be created virtualMachine is concentrated the energy-conservation cloud computing data center virtual machine laying method of placing.
Energy-conservation cloud computing data center virtual machine laying method of the present invention, its special feature is: a. sets up physicsServer zone, all physical servers that an air outlet of air-conditioning refrigeration system is covered are divided into Same Physical serverSubregion, forms the physical server region based on cloud computing; And establish physical server region laterally, the thing that comprises on longitudinallyReason server partition number is respectively x, y, and x, y are positive integer; B. obtain physical server information, institute in obtaining step aThere are resource using information and the place partition information of physical server; C. calculate resources of virtual machine to be created, obtain each waiting and createBuild CPU, internal memory and the hard disk size of virtual machine, and calculate and will create the required request resource of all virtual machines; If requiredThe request resource of CPU, internal memory and hard disk is respectively、With; D. to physical serverSort, all physical servers are sorted according to the size of available resources, form physical server arrangement set, shouldArrangement set is designated as; E. set up the mapping between virtual machine to be created and physical server, meet and wait to create in available resourcesBuild on the basis of virtual machine request resource the physical server resource sequence that search satisfies the demands; And according to inclusion in sequenceReason number of servers number this resource sequence is classified, form physical server arrangement set, be designated as respectively，，…，, whereinRepresent to comprise the set of all sequences that n physical server and available resources satisfy the demands, nFor positive integer; F. judged whether that single physical server satisfies the demands, if there is setExist,In any oneThe available resources of platform physical server all can create out all virtual machines, select wherein arbitrary physical server as void to be createdThe node of plan machine; If do not there is not set, perform step g; G. judge whether to exist meeting the demands in same subregionPhysical server sequence, traversal physical server arrangement set，…，, find out and be positioned at Same Physical server and dividePhysical server sequence in district creates all virtual machines in this physical server sequence; If do not existed in same subregionThe interior physical server sequence meeting the demands, performs step h; H. choose air conditioning energy consumption cost minimization and physical serverThe server sequence of minimum number, creates all virtual machines; Air conditioning energy consumption cost size is large with physical server intensityLittle being inversely proportional to.
In step a, in each physical server subregion, temperature sensor separately should be set, so that air-conditioning refrigeration system rootControl the switching of corresponding air outlet according to the temperature detecting; Physical server subregion is called for short subregion. In step b, physical serverResource using information be the use information of CPU, internal memory and hard disk, place partition information is the seat of physical server place subregionMark information. In step c,For the CPU sum of all virtual machines to be created,For all to be created virtualThe internal memory sum of machine,For the hard disk sum of all virtual machines to be created. In steps d, physical server is arrangedOrder is to form mapping in order to be convenient in step e with virtual machine to be created. In step e, form the property that meets resource requirementReason server arrangement set, so that select the virtual machine placement schemes that energy consumption is minimum. In step f, if there is setExist, show to have the available resources of a physical server can meet the requirement that creates all virtual machines, at a thingOn reason server, create, to reach best energy-conservation object. In step g, on the physical server in same subregionCreate all virtual machines, be also conducive to the concentrated heat radiation of server, reach preferably energy-conservation object. In step h, if there is noThe physical server of one or same subregion meets the requirement that creates virtual machine, chooses the thing in two or more subregionsReason server creates virtual machine, chooses the server sequence of physical server intensity maximum and places virtual machine, to fallLow energy consumption cost.
Energy-conservation cloud computing data center virtual machine laying method of the present invention, establishes the position of physical server subregion with twoDimension coordinate P=(i, j) represent; Air conditioning energy consumption cost indicates with cost, definition:
WhereinFor the number of the subregion that comprises in the physical server resource sequence satisfying the demands,MarkShow the topology distance between physical server place two subregions that satisfy condition;, indicate the thing satisfying conditionThe reason server place subregion summation of the topology distance between subregion between two; It is characterized in that: in described step g, if calculatedThe value that goes out the air conditioning energy consumption cost cost of physical server arrangement set is 1, shows that this physical server arrangement set is positioned atIn Same Physical server partition; If physical server subregion、Coordinate be respectively (i1, j1), (i2, j2),=; In step h, according to，…，Order ask for all physics servicesThe value of the air conditioning energy consumption cost cost of device sequence, chooses the corresponding physical server sequence of first minimum cost value and createsAll virtual machines.
For the computing formula of air conditioning energy consumption cost cost, the number of employing subregion and the between two topology distance between subregionSum indicates, and more can reflect the intensity of physical server in server sequence, can preferably show out actual energyConsumption size. ForAdopt formulaCalculating, is that physical server subregion is at horizontal strokeUnder the condition all equating to, longitudinal pitch, draw.
Energy-conservation cloud computing data center virtual machine laying method of the present invention, in steps d, described physical serverAvailable resources comprise the available resources of CPU, internal memory and hard disk, use respectively、WithMarkShow, adopt respectively following formula to calculate:
Wherein, bottom right is labeled as the expression total resources of total, has used stock number for the expression of used, is thresholdExpression set reserved threshold values; In step e, if the available money of the CPU of one or more physical computers, internal memory and hard diskSource is more than or equal to respectively、With, think that this sequence is the physics clothes that satisfy the demandsBusiness device resource sequence.
In physical server, the use of CPU, internal memory, hard disk being arranged to corresponding reserved threshold values, is to ensure that server is normalThe needs of running, therefore, the quantity available of resource is not known resource total amount and the current simple difference of usage quantity, shouldOn this difference basis, deduct the reserved threshold values of each resources definition.
The invention has the beneficial effects as follows: cloud computing of the present invention data center virtual machine laying method, preferentially select satisfiedThe physical server requiring creates virtual machine; In the case of not existing the physical server meeting the demands, thenSelect the physical server in same subregion to create virtual machine; At the above two, all in non-existent situation, final choice is concentratedThe physical server sequence that degree is the highest creates virtual machine, has realized the most energy-conservation virtual machine laying method. Of the present inventionIn cloud computing data center virtual machine laying method, by setting up air conditioning energy consumption cost cost computing formula, can be efficiently and accuratelyCharacterize the size of air conditioning energy consumption cost by the intensity of physical server, place to select the most energy-conservation virtual machineScheme.
Brief description of the drawings
Fig. 1 is the flow chart of virtual machine laying method of the present invention;
Fig. 2 is the structural representation of physical server subregion in the present invention;
Fig. 3 is that in the present invention, the topology distance between different physical server subregions is calculated schematic diagram.
Detailed description of the invention
Below in conjunction with accompanying drawing and embodiment, the invention will be further described.
As shown in Figure 1, Figure 2 and Figure 3, flow chart, the physical server subregion of virtual machine laying method have been provided respectivelyTopology distance between structural representation and different subregion is calculated schematic diagram, and cloud computing of the present invention data center virtual machine is putThe method of putting comprises the following steps:
A. set up physical server region, all physical servers that an air outlet of air-conditioning refrigeration system is covered are drawnBe divided into Same Physical server partition, form the physical server region based on cloud computing; And establish physical server region at horizontal strokeBe respectively x, y to, the physical server number of partitions that comprises on longitudinally, x, y are positive integer;
If the position of physical server subregion two-dimensional coordinate P=(i, j) represent; Air conditioning energy consumption cost indicates with cost, fixedJustice:
WhereinFor the number of the subregion that comprises in the physical server resource sequence satisfying the demands,MarkShow the topology distance between physical server place two subregions that satisfy condition;, indicate the thing satisfying conditionThe reason server place subregion summation of the topology distance between subregion between two;
B. obtain physical server information, in obtaining step a, the resource using information of all physical servers and place are dividedDistrict's information; The resource using information of physical server is CPU, internal memory and hard disk information, the seat that partition information is physical serverMark information;
C. calculate resources of virtual machine to be created, obtain CPU, internal memory and the hard disk size of each virtual machine to be created, and meterCalculate and will create the required request resource of all virtual machines; If the request resource of required CPU, internal memory and hard disk is respectively、With；
D. physical server is sorted, all physical servers are sorted according to the size of available resources, shapeBecome physical server arrangement set, this arrangement set is designated as；
The available resources of described physical server comprise the available resources of CPU, internal memory and hard disk, use respectively、WithIndicate, adopt respectively following formula to calculate:
Wherein, bottom right is labeled as the expression total resources of total, has used stock number for the expression of used, is thresholdExpression set reserved threshold values;
E. set up the mapping between virtual machine to be created and physical server, meeting virtual machine to be created in available resources pleaseAsk on the basis of resource the physical server resource sequence that search satisfies the demands; And according to comprising physics server count in sequenceAmount number this resource sequence is classified, form physical server arrangement set, be designated as respectively，，…，, whereinRepresent to comprise the set of all sequences that n physical server and available resources satisfy the demands, n is for just wholeNumber;
In this step, if the available resources of the CPU of one or more physical computers, internal memory and hard disk are greater than respectively orEqual、With, think that this sequence is the physical server resource order satisfying the demandsRow;
F. judged whether that single physical server satisfies the demands, if there is setExist,In any oneThe available resources of platform physical server all can create out all virtual machines, select wherein arbitrary physical server as void to be createdThe node of plan machine; If do not there is not set, perform step g;
G. judge whether to exist the physical server sequence meeting the demands in same subregion, traversal physical server sequenceSet，…，, find out the physical server sequence that is positioned at Same Physical server partition, in this physics serviceIn device sequence, create all virtual machines; If there not being the physical server sequence meeting the demands in same subregion, holdRow step h;
In this step, be 1 if calculate the value of the air conditioning energy consumption cost cost of physical server arrangement set, showThis physical server arrangement set is positioned at Same Physical server partition;
H. the server sequence of choosing air conditioning energy consumption cost minimization and physical server minimum number, creates all voidPlan machine; Air conditioning energy consumption cost size is inversely proportional to physical server intensity size;
If physical server subregion、Coordinate be respectively (i1, j1), (i2, j2),=; In step h, according to，…，Order ask for the air-conditioning of all physical server sequencesThe value of energy consumption cost cost, chooses the corresponding physical server sequence of first minimum cost value and creates all virtual machines.
Wherein, described reserved threshold values refers in order to ensure that physical server runs well as the resource reservation of operation on itCertain extending space and the minimum residual capacity of the CPU that sets, internal memory, hard disk. The topology distance of by stages refers to two thingsDistance between reason server partition, for weighing the intensity between subregion.
As shown in Figure 2, be divided into identical subregion in the physical server in the same air outlet of air-conditioning refrigeration system regionIn, in the process that virtual machine is placed, first judge whether that the available resources of a physical server meet establishment requirement,If exist, create virtual machine on this physical server; So just make virtual machine be concentrated, be convenient to unified heat radiation,Realize effective energy-conservation object. If there is no the single physical server satisfying the demands, judges whether to exist in sameThe server sequence that satisfied establishment in one subregion requires, as existed, adopts the server sequence in same subregion to enterRow creates, and is so also conducive to the concentrated of physical server, is convenient to that air-conditioning refrigeration system is same carries out temperature control, is also conducive toThe saving of the energy.
If single physical server recited above or the server sequence in same subregion do not exist,Adopt the size of calculating air conditioning energy consumption cost cost, choose the physical server sequence of the most energy-conservation placement virtual machine. AsShown in Fig. 3, suppose to have two physical servers to meet the requirement that creates all virtual machines, its respectively in subregion Region9,In Region12, for formula,=2，=, cost=5. Also exist 3 physical servers to meet the requirement that creates all virtual machines, wherein 1 is positioned at subregion Region9In, remaining two are arranged in subregion Region12. Owing to being 2 station servers under the first situation, second case is 3 clothesBusiness device, chooses the sequence of physical server minimum number and places virtual machine, to realize best energy-saving effect.
As shown in Figure 3, (establish its coordinate points if comprise 3 physical servers in the satisfied server sequence that creates requirementBe respectively、、), it in subregion Region4, Region9 and Region12, is calculating air conditioning energy consumption one-tenth respectivelyIn the process of this cost,=3，=++=++=5+, so both can calculate air conditioning energy consumptionThe size of cost cost.
As a concrete experimental applications, cloud computing data center topology, as shown in reference to figure 2, has 6 serversSubregion, each subregion is furnished with 4 physical servers, single physical server configuration IntelXeonE56202.4G4 core 8 linesThread processor, 32GBRAM, SAS2.0TRAID5 hard disk, physical server adopts UbuntuServer12.04LTS behaviourTo make system, adopt XenServer5.5 as virtualization software. virtual machine creating adopts unified specification, distributes IntelXeonThe CPU of E5620*1,2G internal memory, 30G hard disk, operation WindowsServer2003SP1 operating system. refrigeration adopts Hai RuiThis (HIRES) precision air conditioner SuperPrecisionHFDC/HFUC0070, main engine power 6.5KW, power of fan 0.25KW.
The virtual machine that test adopts the conventional random virtual machine laying method of current data center and the present invention to propose is placedMethod, places experiment through 10 each 50 virtual machines, adopts the inventive method virtual machine can be concentrated and is placed into 1-2 thingIn reason subregion, and random laying method is generally placed into virtual machine in 3 and above Physical Extents, because virtual machine is placedCan cause the rising of subregion temperature, trigger the work of air-conditioner air outlet, therefore adopt the inventive method can effectively reduce temperatureThe quantity in rising region, thus reduce air conditioning energy consumption, it is estimated under this experimental enviroment, can save air-conditioning electric energy approximately 10% withOn.
Claims (2)
1. an energy-conservation cloud computing data center virtual machine laying method, comprises the following steps:
A. set up physical server region, all physical servers that an air outlet of air-conditioning refrigeration system is covered are divided intoSame Physical server partition, forms the physical server region based on cloud computing; And establish physical server region horizontal, verticalThe physical server number of partitions that upwards comprised is respectively x, y, and x, y are positive integer;
B. obtain physical server information, the resource using information of all physical servers and place subregion letter in obtaining step aBreath;
C. calculate resources of virtual machine to be created, obtain CPU, internal memory and the hard disk size of each virtual machine to be created, and calculateCreate the required request resource of all virtual machines; If the request resource of required CPU, internal memory and hard disk is respectively CPUdemand、RAMdemandAnd DISKdemand；
D. physical server is sorted, all physical servers are sorted according to the size of available resources, formationReason server resource sequence, this resource sequence is designated as PM;
E. set up the mapping between virtual machine to be created and physical server, meet virtual machine request money to be created in available resourcesOn the basis in source, the physical server resource sequence that search satisfies the demands; And according to comprising physics number of servers in sequenceHow many this resource sequence is classified, form physics server resource arrangement set, be designated as respectively VM1，VM2，…，VMn, itsMiddle VMnRepresent to comprise the set of all sequences that n physical server and available resources satisfy the demands, n is positive integer;
F. judged whether that single physical server satisfies the demands, if there is set VM1There is VM1In any physics clothesBusiness device available resources all can create out all virtual machines, select wherein arbitrary physical server as the joint of virtual machine to be createdPoint; If do not there is not set VM1, perform step g;
G. judge whether to exist the physical server resource sequence meeting the demands in Same Physical server partition, traversal physicsServer resource arrangement set VM2，…，VMn, find out the physical server resource order that is positioned at Same Physical server partitionRow create all virtual machines on this physical server resource sequence; As do not existed in Same Physical server partitionThe physical server resource sequence meeting the demands, performs step h;
H. the physical server resource sequence of choosing air conditioning energy consumption cost minimization and physical server minimum number, creates instituteThere is virtual machine; Air conditioning energy consumption cost size is inversely proportional to physical server intensity size;
If the position of physical server subregion represents with two-dimensional coordinate P=(i, j); Air conditioning energy consumption cost represents with cost, definition:
The number that wherein α is the subregion that comprises in the physical server resource sequence satisfying the demands, L (Pu-Pv) represent to meet barTopology distance between physical server place two subregions of part;Represent the physics service satisfying conditionThe device place subregion summation of the topology distance between subregion between two;
It is characterized in that: in described step g, if calculate the air conditioning energy consumption cost cost of physical server resource sequence setValue be 1, show that this physical server resource sequence set is positioned at Same Physical server partition; If physical serverSubregion Pu、PvCoordinate be respectively (iu，ju)、(iv，jv),In step h, pressAccording to VM2，…，VMnOrder ask for the value of the air conditioning energy consumption cost cost of all physics server resource sequences row, choose firstThe corresponding physical server resource sequence of individual minimum cost value creates all virtual machines.
2. energy-conservation cloud computing data center virtual machine laying method according to claim 1, is characterized in that: steps dIn, the available resources of described physical server comprise the available resources of CPU, internal memory and hard disk, use respectively CPUavailable、RAMavailableAnd DISKavailableRepresent, adopt respectively following formula to calculate:
CPUavailable＝CPUtotal-CPUused-CPUthreshold
RAMavailable＝RAMtotal-RAMused-RAMthreshold
DISKavailable＝DISKtotal-DISKused-DISKthreshold
Wherein, bottom right is labeled as the expression total resources of total, has used stock number for the expression of used, is the table of thresholdShow the reserved threshold values of setting;
In step e, if the available resources of the CPU of one or more physical servers, internal memory and hard disk are more than or equal to respectivelyCPUdemand、RAMdemandAnd DISKdemand, think that this sequence is the physical server resource sequence satisfying the demands.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210385961.XA CN102929687B (en) | 2012-10-12 | 2012-10-12 | A kind of energy-conservation cloud computing data center virtual machine laying method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210385961.XA CN102929687B (en) | 2012-10-12 | 2012-10-12 | A kind of energy-conservation cloud computing data center virtual machine laying method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102929687A CN102929687A (en) | 2013-02-13 |
| CN102929687B true CN102929687B (en) | 2016-05-25 |
Family
ID=47644497
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210385961.XA Active CN102929687B (en) | 2012-10-12 | 2012-10-12 | A kind of energy-conservation cloud computing data center virtual machine laying method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102929687B (en) |
Families Citing this family (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103812789A (en) * | 2013-09-18 | 2014-05-21 | 广东电网公司佛山供电局 | Cloud service resource automatic allocating method and system |
| CN103747059B (en) * | 2013-12-26 | 2016-10-05 | 华中科技大学 | A kind of cloud computing server cluster network support method towards many tenants and system |
| CN104063261A (en) * | 2014-04-01 | 2014-09-24 | 杭州电子科技大学 | Multi-objective optimization virtual machine placing method under cloud environment |
| CN104065547B (en) * | 2014-06-23 | 2017-06-20 | 西安电子科技大学昆山创新研究院 | A kind of system of selection for calculating central interior physical host |
| CN104468803B (en) * | 2014-12-12 | 2018-02-09 | 华为技术有限公司 | A kind of virtual data center method for mapping resource and equipment |
| CN104580194B (en) * | 2014-12-31 | 2017-11-24 | 北京四达时代软件技术股份有限公司 | Towards the virtual resources management method and device of Video Applications |
| EP3253027B1 (en) * | 2015-03-03 | 2020-04-29 | Huawei Technologies Co., Ltd. | Resource allocation method and apparatus for virtual machines |
| CN106161539A (en) * | 2015-04-12 | 2016-11-23 | 北京典赞科技有限公司 | Schedule creating energy conservation optimizing method based on the fictitious host computer of ARM server |
| EP3118741A1 (en) | 2015-07-15 | 2017-01-18 | Deutsche Telekom AG | Method, device and system for energy-efficient use of computing units |
| CN106933646B (en) * | 2015-12-29 | 2020-04-14 | 杭州华为数字技术有限公司 | Method and device for creating virtual machine |
| CN107171976B (en) * | 2016-03-07 | 2022-03-25 | 中兴通讯股份有限公司 | Method and device for realizing resource reservation |
| CN105824686B (en) * | 2016-03-11 | 2019-03-22 | 中国联合网络通信集团有限公司 | A kind of selection method of virtual machine host machine and selection system |
| CN105978719B (en) * | 2016-05-10 | 2019-07-16 | 广东睿江云计算股份有限公司 | A kind of control method and device of cloud host services quality |
| CN106095531B (en) * | 2016-06-08 | 2019-06-11 | 北京航空航天大学 | A kind of dispatching method of virtual machine loaded based on grade and physical machine in cloud platform |
| CN106096874A (en) * | 2016-08-18 | 2016-11-09 | 中国联合网络通信集团有限公司 | The energy efficiency managing method of data center, device and system |
| CN108574587B (en) * | 2017-03-09 | 2020-07-24 | 华为技术有限公司 | Capacity updating method and device for distributed equipment |
| CN110795203B (en) * | 2018-08-03 | 2023-05-02 | 阿里巴巴集团控股有限公司 | Resource scheduling method, device, system and computing equipment |
| CN111352721A (en) * | 2018-12-21 | 2020-06-30 | 中国移动通信集团山东有限公司 | Service migration method and device |
| CN110784335B (en) * | 2019-09-19 | 2022-09-02 | 烽火通信科技股份有限公司 | Network element resource reservation system under cloud scene |
| CN110750330A (en) * | 2019-10-18 | 2020-02-04 | 北京浪潮数据技术有限公司 | Virtual machine creating method, system, electronic equipment and storage medium |
| CN111800506B (en) * | 2020-07-06 | 2023-09-19 | 深圳市网心科技有限公司 | Edge computing node deployment method and related device |
| CN113094149B (en) * | 2021-03-12 | 2024-01-12 | 西安交通大学 | Data center virtual machine placement method, system, medium and equipment |
| CN113590342B (en) * | 2021-09-29 | 2022-01-18 | 睿至科技集团有限公司 | Resource allocation method and system in cloud computing system |
| CN114048004A (en) * | 2021-11-22 | 2022-02-15 | 北京志凌海纳科技有限公司 | High-availability batch scheduling method, device, equipment and storage medium for virtual machines |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101556493A (en) * | 2008-04-09 | 2009-10-14 | 株式会社日立制作所 | Operations management methods and devices in information processing systems |
| CN101719081A (en) * | 2009-12-01 | 2010-06-02 | 北京大学 | Method for scheduling virtual machines |
| CN102096461A (en) * | 2011-01-13 | 2011-06-15 | 浙江大学 | Energy-saving method of cloud data center based on virtual machine migration and load perception integration |
| CN102299964A (en) * | 2011-08-30 | 2011-12-28 | 南京邮电大学 | Data redeployment method applied to energy-saving trunking system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8799897B2 (en) * | 2011-04-07 | 2014-08-05 | Vmware, Inc. | Automated cost calculation for virtualized infrastructure |
-
2012
- 2012-10-12 CN CN201210385961.XA patent/CN102929687B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101556493A (en) * | 2008-04-09 | 2009-10-14 | 株式会社日立制作所 | Operations management methods and devices in information processing systems |
| CN101719081A (en) * | 2009-12-01 | 2010-06-02 | 北京大学 | Method for scheduling virtual machines |
| CN102096461A (en) * | 2011-01-13 | 2011-06-15 | 浙江大学 | Energy-saving method of cloud data center based on virtual machine migration and load perception integration |
| CN102299964A (en) * | 2011-08-30 | 2011-12-28 | 南京邮电大学 | Data redeployment method applied to energy-saving trunking system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102929687A (en) | 2013-02-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102929687B (en) | A kind of energy-conservation cloud computing data center virtual machine laying method | |
| Zhou et al. | Minimizing SLA violation and power consumption in Cloud data centers using adaptive energy-aware algorithms | |
| CN106502792B (en) | A kind of multi-tenant priority scheduling of resource method towards different type load | |
| Wang et al. | A new multi-objective bi-level programming model for energy and locality aware multi-job scheduling in cloud computing | |
| CN105159751A (en) | Energy-efficient virtual machine migration method in cloud data center | |
| Samriya et al. | Intelligent SLA-aware VM allocation and energy minimization approach with EPO algorithm for cloud computing environment | |
| Du et al. | Energy-efficient scheduling for tasks with deadline in virtualized environments | |
| Issa | Performance evaluation and estimation model using regression method for hadoop wordcount | |
| CN106775987A (en) | A kind of dispatching method of virtual machine for improving resource efficiency safely in IaaS cloud | |
| CN108874508A (en) | A kind of cloud computing virtual server system load equilibration scheduling method | |
| Tuli et al. | Efficient virtual machine migration algorithms for data centers in cloud computing | |
| Biswas et al. | A novel resource aware scheduling with multi-criteria for heterogeneous computing systems | |
| Giacobbe et al. | An approach to reduce energy costs through virtual machine migrations in cloud federation | |
| CN106844175B (en) | A kind of cloud platform method for planning capacity based on machine learning | |
| Cordsen et al. | Object-oriented operating systems design and the revival of program families | |
| Quang-Hung et al. | Heuristics for energy-aware VM allocation in HPC clouds | |
| Ali et al. | FollowMe@ LS: Electricity price and source aware resource management in geographically distributed heterogeneous datacenters | |
| Dhanoa et al. | Power efficient hybrid VM allocation algorithm | |
| Kumar et al. | Power and data aware best fit algorithm for energy saving in cloud computing | |
| Teodoro et al. | Energy efficiency management in computational grids through energy-aware scheduling | |
| Sharma et al. | Impact of energy-efficient and eco-friendly green computing | |
| Goyal et al. | Energy optimised resource scheduling algorithm for private cloud computing | |
| Merzoug et al. | Intelligent strategy of allocation resource for cloud datacenter based on mas & cp approach | |
| Hage et al. | Saving the planet with bin packing-Experiences using 2D and 3D bin packing of virtual machines for greener clouds | |
| Kumar et al. | Energy Efficient Model for Balancing Energy in Cloud Datacenters Using Dynamic Voltage Frequency Scaling (DVFS) Technique |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: 250014 Ji'nan Province, Shandong Province, Lixia District, Department of road, No. 19, Shandong Computing Center Applicant after: SHANDONG COMPUTER SCIENCE CENTER Address before: 250014 Ji'nan Province, Shandong Province, Lixia District, Department of road, No. 19, Shandong Computing Center Applicant before: Shandong Prov. Computing Center |
|
| COR | Change of bibliographic data | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |