CN105426241A - Cloud computing data center based unified resource scheduling energy-saving method - Google Patents
Cloud computing data center based unified resource scheduling energy-saving method Download PDFInfo
- Publication number
- CN105426241A CN105426241A CN201510782854.4A CN201510782854A CN105426241A CN 105426241 A CN105426241 A CN 105426241A CN 201510782854 A CN201510782854 A CN 201510782854A CN 105426241 A CN105426241 A CN 105426241A
- Authority
- CN
- China
- Prior art keywords
- virtual machine
- node
- resource
- network node
- energy consumption
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/485—Task life-cycle, e.g. stopping, restarting, resuming execution
- G06F9/4856—Task life-cycle, e.g. stopping, restarting, resuming execution resumption being on a different machine, e.g. task migration, virtual machine migration
- G06F9/4862—Task life-cycle, e.g. stopping, restarting, resuming execution resumption being on a different machine, e.g. task migration, virtual machine migration the task being a mobile agent, i.e. specifically designed to migrate
- G06F9/4875—Task life-cycle, e.g. stopping, restarting, resuming execution resumption being on a different machine, e.g. task migration, virtual machine migration the task being a mobile agent, i.e. specifically designed to migrate with migration policy, e.g. auction, contract negotiation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F1/00—Details not covered by groups G06F3/00 - G06F13/00 and G06F21/00
- G06F1/26—Power supply means, e.g. regulation thereof
- G06F1/32—Means for saving power
- G06F1/3203—Power management, i.e. event-based initiation of a power-saving mode
- G06F1/3234—Power saving characterised by the action undertaken
- G06F1/329—Power saving characterised by the action undertaken by task scheduling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/60—Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources
-
- 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
The invention discloses a cloud computing data center based unified resource scheduling energy-saving method. The method comprises the following steps: 1, initializing a network node and a virtual machine queue; 2, storing a virtual machine request in the virtual machine queue; 3, arranging virtual machines in a descending order according to a resource request number of the virtual machines; 4, traversing all network nodes in sequence, and judging whether a network node meets a request requirement of a current virtual machine or not; if yes, taking a network node with lowest energy consumption, requested by the current virtual machine, as a target placement node, and otherwise, looking for a network node with most residual available resources, emigrating one virtual machine on the node, and placing the current virtual machine; 5, sequentially selecting a next virtual machine as a current virtual machine, and making a judgment again; and 6, optimizing system energy consumption again. The method has the advantages that a contradictory relation between a power consumption minimization problem and SLA requirement satisfaction is balanced; the method is an energy consumption optimization oriented resource scheduling algorithm; and the method has higher efficiency in energy consumption optimization.
Description
Technical field
The invention belongs to field of cloud calculation, specifically a kind of scheduling of the unified resource based on cloud computation data center power-economizing method.
Background technology
Along with proposition and the development of cloud computing, the high energy consumption problem of data center starts the extensive concern receiving various circles of society, a large amount of computational resources and storage resources are transitioned into from distributed loose placement and concentrate on high in the clouds, bring larger challenge to the efficient management of energy consumption.Because the data volume of cloud computation data center increases day by day, the deviser of cloud computing infrastructure (IaaS) and supvr are faced with huge fiscal stimulus and regulation and control pressure, wish to find effective strategy to reduce the operation and maintenance cost of cloud computing system, reduce the energy consumption of cloud computation data center, and then reduce CO2 emission.
Current Researching and practicing takes two kinds of methods to realize efficiency optimization usually: one is that data center adopts regenerative resource to power; Another kind method is the polymerization and the migration that utilize virtual machine, the virtual machine of integration work, improves the service efficiency of physical resource, reduces energy resource consumption.
The various saving energy consuming process in cloud computing architecture are are much researched and proposed.Wherein most of method lays particular emphasis on the distribution of virtual machine and the Resource Management Algorithm of scheduling with energy consumption perception (EnergyAware) characteristic.Its basic thought is: reduce the number of servers distributing to Virtual Machine Worker, closes or those idle machines without any work of dormancy, thus reaches the efficiency improving operational outfit, and then in cloud computing infrastructure layer, save the object of energy consumption.In addition, can also, by virtual machine is placed on the server of minimum number, make part computational resource be transitioned into low power consumpting state, improve the utilization factor of resource.
Operating loads different in cloud computing environment is also consumption of data center problem facing challenges, and user can run various types of task, as social networks, scientific algorithm, commerce services.These tasks, often in processor utilization rate, internal memory and memory access, show different features.Meanwhile, all these different operating loads, usually on service-level agreement, have different quality of service requirements.This resource scheduling algorithm making design one outstanding, becomes complex.
At present, in cloud computation data center, the polymerization of resources of virtual machine and migration, can be considered an optimization problem having Multiple Optimization target and constraint condition.In order to find optimal solution, the distribution of virtual machine and transition process, the example of a multidimensional Packing Problems (BinPackingProblem) can be regarded as.The difficult character of NP-due to this problem, the heuristic of people verified such as " First-Fit-Decreasing-Resource " and " Best-Fit-Decreasing-Resource " can provide good effect for this problem.In order to assess different heuritic approaches, researcher uses the mode of experiment or emulation usually, tests these algorithms effect in different situations.But in real experimental situation, be the energy consumption of an accurate analysis data center, be necessary for it and set up a large amount of energy consumption measurement probes.Meanwhile, in order to performance and the consequence of test resource dispatching algorithm, expensive energy is also needed to remove the resource dispatching strategy at Update Table center.Therefore, simulation becomes in extensive cloud computation data center, and research has the most feasible way of the resource scheduling algorithm of energy optimization.
Summary of the invention
The present invention is directed in prior art, the problems such as the high and energy-saving effect of the complexity that in the data heart scheduling of resource managing power consumption exists is not obvious, propose a kind of unified resource based on cloud computation data center and dispatch power-economizing method.
Comprise the following steps
The network node that step one, initialization data center are all and virtual machine queue;
Each network node of initialization comprises: 1) use of the energy consumption of each node, resource are used and be set to zero, comprise and CPU and internal memory are set to zero; 2) be the upper lower threshold value of utilization rate setting of each network node;
Virtual machine list collection is set to empty set;
Step 2, by user propose virtual machine request be stored in virtual machine queue;
Step 3, the virtual machine request submitted to according to user, to the resource request number descending sort of virtual machine according to virtual machine.
The virtual machine many for resource request number is preferentially placed.
Step 4, for the virtual machine after arrangement, travel through all-network node successively, judged whether that network node can meet current virtual machine request.If met, enter step 5; Otherwise enter step 6;
Current virtual machine is designated as Vm
j, 1≤j≤n, n is the sum of the virtual machine of request resource; Vm
jinitial value is the virtual machine that resource request number is maximum;
Step 5, obtain and meet the minimum network node of the energy consumption of current virtual machine request as drop target node, enter step 7;
Be specially:
Step 501, successively traversal all-network node, mark meeting current virtual machine number of resource requests object network node;
If the available resources number of certain network node is more than or equal to the resource request number of current virtual machine, then this network node meets virtual machine request, and Boolean variable W (Node) value marking this network node is 1;
Step 502, each network node for mark, calculate virtual machine respectively and be placed on energy consumption on each token network node;
Formula ENERGY-ESTIMATE is used to calculate;
Step 503, choose the minimum network node of energy consumption as drop target node;
Step 504, current virtual machine to be placed on drop target node, and to upgrade drop target nodal information, enter step 7.
Step 6, again traversal all-network node, obtain the network node that remaining available resource is maximum, certain virtual machine of moving out on this node, place current virtual machine, enter step 7;
Be specially:
Step 601, traversal all-network node, find the network node that remaining available resource is maximum, be designated as Node
k
Step 602, to network node Node
kthe use resource ascending order arrangement of upper each virtual machine;
Step 603, successively choose from small to large sequence after virtual machine as virtual machine Vm to be moved out
t;
Virtual machine Vm to be moved out
tinitial value is node Node
kthe virtual machine that upper use resource is minimum;
Step 604, judge will virtual machine Vm be moved out
tafter moving out, node Node
kremaining available resource can meet current virtual machine Vm
jresource request, if can, will virtual machine Vm be moved out
tmove out, simultaneously by current virtual machine Vm
jbe placed into node Node
k, and more new node Node
kinformation.Otherwise, current virtual machine Vm
jsuccessfully do not place, return step 603.
Step 605, traveling through all the other network nodes, is virtual machine Vm to be moved out
tfind destination node to place;
Calculate the current residual available resources of all the other each network nodes, searching can meet virtual machine Vm
tresource request, and the node that resources of virtual machine utilization rate is the highest, will virtual machine Vm be moved out
tmove to this node, and upgrade this nodal information.
If step 606 network node Node
k, after moving out using the maximum virtual machine of resource, remaining available resources still do not meet current virtual machine Vm
jresource request number, then return step 601 continue traversal all the other all-network nodes.
Step 7, order choose next virtual machine in virtual machine queue as current virtual machine, return step 4; Until place complete by virtual machines all in virtual machine queue, algorithm terminates.
Step 8, resource extent according to the actual use of virtual machine, optimize system energy consumption again.
After virtual machines all in virtual machine queue are placed, when not having new virtual machine request to arrive in cluster, by formula ENERGY-ESTIMATE, the current energy consumption of each node is calculated to each network node in cluster;
Be specially: first the use resource ascending order of virtual machine on each node is arranged; Then the remaining available resource of each node is calculated; If the remaining available resource on certain network node Node meets the use resource of certain virtual machine Vm on another network node, then virtual machine Vm is moved on network node Node, upgrade the information of network node Node, the use resource of network node Node is maximized;
Or after being used by certain network node the less virtual machine of resource to move out, the remaining available resource of this network node meets resource and uses the virtual machine that number is larger, by virtual machine (vm) migration less for resource, place resource and use the larger virtual machine of number and upgrade this network node information.
By all-network node through migration process; More surplus resources is discharged for next virtual machine request queue after optimizing.
The invention has the advantages that:
1, based on a unified resource scheduling power-economizing method for cloud computation data center, by carrying out modeling to the energy consumption problem of cloud computation data center, this mathematical model have total system, accurately and fast, versatility, elasticity, the feature such as simple.
2, based on a unified resource scheduling power-economizing method for cloud computation data center, be the resource scheduling algorithm towards energy optimization, there is higher efficiency in energy optimization.
3, based on a unified resource scheduling power-economizing method for cloud computation data center, balance minimizing power dissipation problem and meet the contradictory relation between SLA requirement.
Accompanying drawing explanation
Fig. 1 is a kind of scheduling of the unified resource based on cloud computation data center of the present invention power-economizing method process flow diagram;
Fig. 2 is the method flow diagram that the present invention obtains that destination node places virtual machine;
Fig. 3 is that the present invention utilizes virtual machine (vm) migration algorithm to place the method flow diagram of virtual machine.
Embodiment
Below in conjunction with accompanying drawing, the present invention is described in further detail.
Carry out modeling for the energy consumption in cloud computing, design rational dispatching algorithm under the prerequisite not reducing the grade of service, reduce consumption of data center.
Based on a unified resource scheduling power-economizing method for cloud computation data center, input the virtual machine of application, clustered node and current virtual machine list, the resource pool after upgrading and virtual machine list will be exported by computing.First initialization network node and virtual machine queue, next place and migration virtual machine, the placement of virtual machine and migration can change the state of system and node.In order to avoid influencing each other, placement and the migration of virtual machine will separately be carried out.When needing to distribute the virtual machine in virtual machine list, system performs virtual machine Placement, chooses most suitable node placement virtual machine; When virtual machine list is empty, and after continue for T time, system carries out virtual machine (vm) migration operation.
As shown in Figure 1, concrete steps are as follows:
The network node that step one, initialization data center are all and virtual machine queue.
The effect of cloud computation data center network node realizes transfer, load balancing, divides the function such as resource, redundancy equally;
Usually comprise a lot of network nodes in cluster, each network node is defined as different virtual machine operations, and namely each network node is provided with multiple virtual machine;
Virtual machine is the basic thread of cloud computing resources, and the number of virtual machine asks setting according to user.
Initialization operation is as follows: 1), travel through all network nodes, uses (CPU, internal memory etc.) to be set to zero the use of the energy consumption of each network node, resource; Set the upper and lower threshold value of each network node utilization rate.
The object of setting utilization rate threshold value is to improve energy efficiency.By the resource situation of Controlling vertex in the scope of specifying, more resource can be saved, guarantee the service quality of node simultaneously.
2), virtual machine list collection is set to empty set;
Step 2, by user propose virtual machine request be stored in virtual machine queue.
After the network node that initialization uses, virtual machine queue empties.When user submits virtual machine request to, then can be stored in virtual machine queue.Then, be the upper lower threshold value of each network node setting, more total time is set as zero.
Step 3, the virtual machine request submitted to according to user, according to the resources of virtual machine request number descending sort in virtual machine list.
To the scale descending sort of the request in virtual machine list according to application, the resource request number of virtual machine is arranged from big to small.Object is algorithm optimization, and the virtual machine many for resource occupation is preferentially placed.
Step 4, for the virtual machine after arrangement, travel through all-network node successively, judge whether certain network node can meet current virtual machine resource request.If met, enter step 5; Otherwise enter step 6;
Current virtual machine is designated as Vm
j, 1≤j≤n; N is the sum of the virtual machine of request resource; Vm
jinitial value is first virtual machine after arrangement, the virtual machine that namely resource request number is maximum;
Step 5, obtain and meet the minimum network node of the energy consumption of current virtual machine request as drop target node, enter step 7;
As shown in Figure 2, concrete steps are:
Step 501, successively traversal all-network node, mark meeting current virtual machine number of resource requests object network node;
If the available resources number of certain network node is more than or equal to the resource request number of current virtual machine, then this network node meets virtual machine request, and Boolean variable W (Node) value marking this network node is 1;
Step 502, each network node for mark in step 501, calculate virtual machine respectively and be placed on energy consumption on each token network node;
Statistics Boolean variable W (Node) value is the network node of 1, and is placed on the energy consumption on each node by formula ENERGY-ESTIMATE calculating virtual machine;
ENERGY-ESTIMATE in virtual machine Placement is the energy consumption model based on node, and the resource extent of the using state current to server and virtual machine application is relevant.After the energy consumption of virtual machine to server is estimated as and places this virtual machine, the part that server energy consumption increases.
The network node that in step 503, selecting step 502, energy consumption is minimum is as drop target node;
Step 504, current virtual machine to be placed on drop target node, and to upgrade drop target nodal information, enter step 7.
The target of this algorithm is not only and is allowed virtual machine and physical node mate, and the energy consumption of cluster also will be allowed minimum.
From virtual machine list, select a virtual machine i, add in the virtual machine set of drop target node, and virtual machine i is placed on this node, upgrade this nodal information.
Step 6, again traversal all-network node, obtain the network node that remaining available resource is maximum, certain virtual machine of moving out on this node, place current virtual machine, enter step 7;
As shown in Figure 3, be specially:
Step 601, traversal all-network node, find the current network node that surplus resources is maximum, be designated as Node
k
Step 602, to node Node
kthe use resource ascending order arrangement of upper each virtual machine,
On network node, the use resource of virtual machine is less, then the surplus resources of this node is larger;
Step 603, by using the ascending order of resource to choose virtual machine as virtual machine vm to be moved out
t;
Virtual machine Vm to be moved out
tinitial value is node Node
kthe virtual machine that upper use resource is minimum;
Step 604, to judge virtual machine Vm
tafter moving out, node Node
kremaining available resource can meet current virtual machine Vm
jresource request, if meet, will virtual machine Vm be moved out
tmove out, simultaneously by current virtual machine Vm
jbe placed into node Node
k, and more new node Node
kinformation; Otherwise, current virtual machine Vm
jsuccessfully do not place, return step 603.
Step 605, traveling through all the other network nodes, is virtual machine Vm to be moved out
tfind destination node to place;
Calculate the current residual available resources of all the other each network nodes, searching can meet virtual machine Vm
tresource request, and the node that resources of virtual machine utilization rate is the highest, will virtual machine Vm be moved out
tmove to this node, and upgrade this nodal information.
If step 606 network node Node
k, after moving out using the maximum virtual machine of resource, remaining available resources still do not meet current virtual machine Vm
jresource request number, then return step 601 continue traversal all the other all-network nodes;
Step 7, order choose next virtual machine in virtual machine queue as current virtual machine, return step 4; Until place complete by virtual machines all in virtual machine queue, algorithm terminates.
Step 8, resource extent according to the actual use of virtual machine, optimize system energy consumption again.
Use virtual machine (vm) migration algorithm, in virtual machine Placement, virtual machine optimization placement node is the scale based on virtual machine application resource.When system is after operation a period of time tends towards stability, the resource extent of the actual use of virtual machine can be adopted, system energy consumption is optimized again.Virtual machine Placement and virtual machine (vm) migration algorithm are all the subalgorithms of cluster resource dispatching algorithm.
After virtual machines all in virtual machine queue are placed, when not having new virtual machine request to arrive in cluster, by formula ENERGY-ESTIMATE, the current energy consumption of each node is calculated to each node in cluster, predict the energy consumption after virtual machine (vm) migration to other nodes simultaneously, find energy consumption minimum value.If the minimum node of energy consumption is not existing node on this position, replaces existing node with the node that energy consumption is minimum, complete the migration of virtual machine, and upgrade the information of these two nodes.Repeat until all nodes are all through process.More surplus resources is discharged for next virtual machine request queue after optimizing.
Virtual machine (vm) migration algorithm can be counted as the double optimization of the energy consumption to cloud computing cluster.In virtual machine Placement, virtual machine optimization placement node is the scale based on virtual machine application resource.When system is after operation a period of time tends towards stability, the resource extent of the actual use of virtual machine can be adopted, system energy consumption is optimized again.Virtual machine Placement and virtual machine (vm) migration algorithm are all the subalgorithms of cluster resource dispatching algorithm.
Claims (4)
1., based on a unified resource scheduling power-economizing method for cloud computation data center, it is characterized in that, comprise the following steps:
The network node that step one, initialization data center are all and virtual machine queue;
Initialization network node comprises: 1) use of the energy consumption of each node, resource are used and be set to zero; 2) be the upper lower threshold value of utilization rate setting of each network node;
Virtual machine list collection is set to empty set;
Step 2, by user propose virtual machine request be stored in virtual machine queue;
Step 3, the virtual machine request submitted to according to user, to the resource request number descending sort of virtual machine according to virtual machine;
Step 4, for the virtual machine after arrangement, travel through all-network node successively, judged whether that network node can meet current virtual machine request; If met, enter step 5; Otherwise enter step 6;
Current virtual machine is designated as Vm
j, 1≤j≤n, n is the sum of the virtual machine of request resource; Vm
jbe initially the virtual machine that resource request number is maximum;
Step 5, obtain and meet the minimum network node of the energy consumption of current virtual machine request as drop target node, enter step 7;
Step 6, again traversal all-network node, obtain the network node that remaining available resource is maximum, certain virtual machine of moving out on this node, place current virtual machine, enter step 7;
Step 7, order choose next virtual machine in virtual machine queue as current virtual machine, return step 4; Until place complete by virtual machines all in virtual machine queue, algorithm terminates;
Step 8, resource extent according to the actual use of virtual machine, optimize system energy consumption again.
2. a kind of unified resource based on cloud computation data center dispatches power-economizing method as claimed in claim 1, and it is characterized in that, described step 5, is specially:
Step 501, successively traversal all-network node, mark meeting current virtual machine number of resource requests object network node;
If the available resources number of certain network node is more than or equal to the resource request number of current virtual machine, then this network node meets virtual machine request, and Boolean variable W (Node) value marking this network node is 1;
Step 502, each network node for mark, calculate virtual machine respectively and be placed on energy consumption on each token network node;
Step 503, choose the minimum network node of energy consumption as drop target node;
Step 504, current virtual machine to be placed on drop target node, and to upgrade drop target node.
3. a kind of unified resource based on cloud computation data center dispatches power-economizing method as claimed in claim 1, and it is characterized in that, described step 6, is specially: be specially:
Step 601, traversal all-network node, find the network node that remaining available resource is maximum, be designated as Node
k
Step 602, to network node Node
kthe use resource ascending order arrangement of upper each virtual machine;
Step 603, successively choose from small to large sequence after virtual machine as virtual machine Vm to be moved out
t;
Step 604, judge will virtual machine Vm be moved out
tafter moving out, node Node
kremaining available resource can meet current virtual machine Vm
jresource request, if can, will virtual machine Vm be moved out
tmove out, simultaneously by current virtual machine Vm
jbe placed into node Node
k, and more new node Node
kenergy consumption uses and resource using information; Otherwise, current virtual machine Vm
jsuccessfully do not place, return step 603;
Step 605, traveling through all the other network nodes, is virtual machine Vm to be moved out
tfind destination node to place;
Calculate the current residual available resources of all the other each network nodes, searching can meet virtual machine Vm
tresource request, and the node that resources of virtual machine utilization rate is the highest, will virtual machine Vm be moved out
tmove to this node, and upgrade the use of this node energy consumption and resource using information;
If step 606 network node Node
k, after moving out using the maximum virtual machine of resource, remaining available resources still do not meet current virtual machine Vm
jresource request number, then return step 601 continue traversal all the other all-network nodes.
4. a kind of unified resource based on cloud computation data center dispatches power-economizing method as claimed in claim 1, it is characterized in that, described step 8, be specially: after virtual machines all in virtual machine queue are placed, when not having new virtual machine request to arrive in cluster, current energy consumption is calculated respectively to each network node in cluster;
The use resource ascending order of virtual machine on each node is arranged, calculates the remaining available resource of each node simultaneously; If the remaining available resource on certain network node Node meets the use resource of certain virtual machine Vm on another network node, then virtual machine Vm is moved on network node Node, the energy consumption upgrading network node Node uses and resource using information, and the use resource of network node Node is maximized; Or after being used by certain network node the less virtual machine of resource to move out, the remaining available resource of this network node meets resource and uses the virtual machine that number is larger, by virtual machine (vm) migration less for resource, place resource and use the larger virtual machine of number and upgrade the use of this network node energy-consumption and resource using information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510782854.4A CN105426241A (en) | 2015-11-16 | 2015-11-16 | Cloud computing data center based unified resource scheduling energy-saving method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510782854.4A CN105426241A (en) | 2015-11-16 | 2015-11-16 | Cloud computing data center based unified resource scheduling energy-saving method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105426241A true CN105426241A (en) | 2016-03-23 |
Family
ID=55504467
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510782854.4A Pending CN105426241A (en) | 2015-11-16 | 2015-11-16 | Cloud computing data center based unified resource scheduling energy-saving method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105426241A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106201721A (en) * | 2016-07-12 | 2016-12-07 | 浪潮(北京)电子信息产业有限公司 | A kind of internal memory dynamic adjusting method based on Intel Virtualization Technology and system |
| CN106339259A (en) * | 2016-08-15 | 2017-01-18 | 上海欧网网络科技发展有限公司 | Real-time scheduling method for cloud calculation resource |
| CN107301092A (en) * | 2016-04-15 | 2017-10-27 | 中移(苏州)软件技术有限公司 | A kind of cloud computing resource pool energy saving of system method, apparatus and system |
| CN107341043A (en) * | 2017-06-28 | 2017-11-10 | 东北大学 | A kind of emulation mode for the consumption of data center for assessing regenerative resource hybrid power supply |
| CN107733701A (en) * | 2017-09-29 | 2018-02-23 | 中国联合网络通信集团有限公司 | A kind of method and apparatus of deploying virtual machine |
| CN108304261A (en) * | 2017-12-29 | 2018-07-20 | 曙光信息产业(北京)有限公司 | A kind of job scheduling method and device based on 6D-Torus networks |
| CN108415763A (en) * | 2018-02-11 | 2018-08-17 | 中南大学 | A kind of distribution method of edge calculations system |
| CN109086120A (en) * | 2018-08-03 | 2018-12-25 | 北京奇虎科技有限公司 | Management method, device and the calculating equipment of resources of virtual machine |
| CN109992418A (en) * | 2019-03-25 | 2019-07-09 | 华南理工大学 | The multi-tenant big data platform resource priority level scheduling method and system of SLA perception |
| CN111488052A (en) * | 2020-04-16 | 2020-08-04 | 中国工商银行股份有限公司 | Container enabling method and device applied to physical machine cluster and computer system |
| CN116149798A (en) * | 2023-04-13 | 2023-05-23 | 北京中电德瑞电子科技有限公司 | Virtual machine control method and device of cloud operating system and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102404412A (en) * | 2011-12-28 | 2012-04-04 | 北京邮电大学 | Energy saving method and system for cloud compute data center |
| CN102567076A (en) * | 2011-12-12 | 2012-07-11 | 华中科技大学 | Selection method of server integration scheme based on energy consumption |
| US20120198247A1 (en) * | 2010-11-05 | 2012-08-02 | International Business Machines Corporation | Energy Allocation to Groups of Virtual Machines |
-
2015
- 2015-11-16 CN CN201510782854.4A patent/CN105426241A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120198247A1 (en) * | 2010-11-05 | 2012-08-02 | International Business Machines Corporation | Energy Allocation to Groups of Virtual Machines |
| CN102567076A (en) * | 2011-12-12 | 2012-07-11 | 华中科技大学 | Selection method of server integration scheme based on energy consumption |
| CN102404412A (en) * | 2011-12-28 | 2012-04-04 | 北京邮电大学 | Energy saving method and system for cloud compute data center |
Non-Patent Citations (1)
| Title |
|---|
| FEI ZHANG, YIZHOU YAN: "A Heuristics Approach for Reducing Power Consumption of Cloud Data Center", 《2013 IEEE INTERNATIONAL CONFERENCE ON GREEN COMPUTING AND COMMUMCATIONS AND IEEE INTERNET OF THINGS AND IEEE CYBER PHYSICAL AND SOCIAL COMPUTING》 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107301092A (en) * | 2016-04-15 | 2017-10-27 | 中移(苏州)软件技术有限公司 | A kind of cloud computing resource pool energy saving of system method, apparatus and system |
| CN106201721B (en) * | 2016-07-12 | 2019-05-10 | 浪潮(北京)电子信息产业有限公司 | A kind of memory dynamic adjusting method and system based on virtualization technology |
| CN106201721A (en) * | 2016-07-12 | 2016-12-07 | 浪潮(北京)电子信息产业有限公司 | A kind of internal memory dynamic adjusting method based on Intel Virtualization Technology and system |
| CN106339259A (en) * | 2016-08-15 | 2017-01-18 | 上海欧网网络科技发展有限公司 | Real-time scheduling method for cloud calculation resource |
| CN106339259B (en) * | 2016-08-15 | 2019-08-27 | 上海欧网网络科技发展有限公司 | The real-time scheduling method of cloud computing resources |
| CN107341043A (en) * | 2017-06-28 | 2017-11-10 | 东北大学 | A kind of emulation mode for the consumption of data center for assessing regenerative resource hybrid power supply |
| CN107733701A (en) * | 2017-09-29 | 2018-02-23 | 中国联合网络通信集团有限公司 | A kind of method and apparatus of deploying virtual machine |
| CN107733701B (en) * | 2017-09-29 | 2019-11-08 | 中国联合网络通信集团有限公司 | A kind of method and apparatus of deploying virtual machine |
| CN108304261A (en) * | 2017-12-29 | 2018-07-20 | 曙光信息产业(北京)有限公司 | A kind of job scheduling method and device based on 6D-Torus networks |
| CN108304261B (en) * | 2017-12-29 | 2022-05-24 | 曙光信息产业(北京)有限公司 | Job scheduling method and device based on 6D-Torus network |
| CN108415763A (en) * | 2018-02-11 | 2018-08-17 | 中南大学 | A kind of distribution method of edge calculations system |
| CN108415763B (en) * | 2018-02-11 | 2020-12-11 | 中南大学 | Distribution method of edge computing system |
| CN109086120A (en) * | 2018-08-03 | 2018-12-25 | 北京奇虎科技有限公司 | Management method, device and the calculating equipment of resources of virtual machine |
| CN109992418A (en) * | 2019-03-25 | 2019-07-09 | 华南理工大学 | The multi-tenant big data platform resource priority level scheduling method and system of SLA perception |
| CN109992418B (en) * | 2019-03-25 | 2023-01-06 | 华南理工大学 | SLA-aware resource priority scheduling method and system for multi-tenant big data platform |
| CN111488052A (en) * | 2020-04-16 | 2020-08-04 | 中国工商银行股份有限公司 | Container enabling method and device applied to physical machine cluster and computer system |
| CN116149798A (en) * | 2023-04-13 | 2023-05-23 | 北京中电德瑞电子科技有限公司 | Virtual machine control method and device of cloud operating system and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105426241A (en) | Cloud computing data center based unified resource scheduling energy-saving method | |
| CN106534318B (en) | A kind of OpenStack cloud platform resource dynamic scheduling system and method based on flow compatibility | |
| CN102932279B (en) | A kind of cloud environment data center multi dimensional resource dispatching method | |
| CN106020933B (en) | Cloud computing dynamic resource scheduling system and method based on ultralight amount virtual machine | |
| CN104735095B (en) | A kind of cloud computing platform job scheduling method and device | |
| CN106506657A (en) | One kind distributes method of adjustment based on multiobject cloud computing virtual machine | |
| CN103605578B (en) | Load equilibration scheduling method based on virtual machine (vm) migration | |
| CN105446816B (en) | A kind of energy optimization dispatching method towards heterogeneous platform | |
| CN106598733A (en) | Three-dimensional virtual resource scheduling method of cloud computing energy consumption key | |
| CN105610715B (en) | A kind of cloud data center multi-dummy machine migration scheduling method of planning based on SDN | |
| CN106951059A (en) | Based on DVS and the cloud data center power-economizing method for improving ant group algorithm | |
| CN102404412A (en) | Energy saving method and system for cloud compute data center | |
| CN104731528B (en) | A kind of construction method and system of cloud computing block storage service | |
| CN105471985A (en) | Load balance method, cloud platform computing method and cloud platform | |
| CN104023042B (en) | Cloud platform resource scheduling method | |
| CN108519919A (en) | A method of realizing server resource dynamic dispatching under virtual cluster environment | |
| CN109271257A (en) | A kind of method and apparatus of virtual machine (vm) migration deployment | |
| CN110347515A (en) | A kind of resource optimal distribution method of suitable edge calculations environment | |
| CN110311965A (en) | Method for scheduling task and system under a kind of cloud computing environment | |
| Li et al. | Cost-aware automatic scaling and workload-aware replica management for edge-cloud environment | |
| CN109976879B (en) | Cloud computing virtual machine placement method based on resource usage curve complementation | |
| Zhang et al. | An energy-aware host resource management framework for two-tier virtualized cloud data centers | |
| CN108595255A (en) | Workflow task dispatching method based on shortest path first in geographically distributed cloud | |
| CN106407007A (en) | Elasticity analysis process oriented cloud resource allocation optimization method | |
| CN105049499A (en) | Multi-cube mapping-based resource allocation method in network function virtualization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160323 |
|
| RJ01 | Rejection of invention patent application after publication |