CN104052820A - Dynamic energy-saving resource scheduling system and method for distributed cloud computing platform - Google Patents
Dynamic energy-saving resource scheduling system and method for distributed cloud computing platform Download PDFInfo
- Publication number
- CN104052820A CN104052820A CN201410301191.5A CN201410301191A CN104052820A CN 104052820 A CN104052820 A CN 104052820A CN 201410301191 A CN201410301191 A CN 201410301191A CN 104052820 A CN104052820 A CN 104052820A
- Authority
- CN
- China
- Prior art keywords
- resource
- virtual machine
- module
- energy consumption
- computing platform
- 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
-
- 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
Landscapes
- Power Sources (AREA)
Abstract
The invention provides a dynamic energy-saving resource scheduling system and method for a distributed cloud computing platform. The system comprises a cloud server node of the distributed cloud computing platform. The cloud server node is respectively connected with a global controller and a monitoring system; the global controller and the monitoring system are connected with a modeling analysis module respectively; the monitoring system comprises an energy consumption monitoring module, a flow monitoring module and a resource monitoring module; a node controller operates on the cloud server node. The system and method aim at cross-domain distribution characteristics and comprehensive energy-saving resource scheduling demands of the distributed cloud computing platform, and the effect of reducing energy consumption of a server and the network at the same time is achieved.
Description
Technical field
The present invention relates to the system and method in a kind of cloud computing technology field, be specifically related to a kind of dynamic energy-saving resource scheduling system and method for distributed cloud computing platform.
Background technology
Cloud computing platform provides three class cloud services to user: infrastructure serve, platform serves, software serve, the base support Wei Yun data center of these cloud services.In the operation of cloud computing data center, physical server cluster by One's name is legion directly provides virtual machine or member other cloud services on virtual machine is provided for user, the establishment of virtual machine can make physical server cluster produce a large amount of heats, reach as high as degree more than 50, the cost of bringing is that the refrigerating system power consumption that maintains data center's normal temperature increases.Google has 36Ge data center in the whole world, within 1 year, consume about 79,000,000,000 kilowatt 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.Visible, along with cloud data center scale increases day by day, power saving has become the operation of cloud data center and has been badly in need of a great problem solving.Because the core of cloud data center is magnanimity virtual machine, therefore, how rationally to dispose virtual machine to reduce the key problem that energy consumption Cheng Ze is current science and IT enterprises research.
Most of existing virtual machine deployment method is only considered one-sided factor, as minimizes network energy consumption, realizes heat radiation balanced, or maximum resource service efficiency.For example, someone has proposed a kind of delivery flow rate perception deploying virtual machine model TVMPP, and this model can reduce frequent data exchange and the very high data center network energy consumption of bandwidth utilization rate effectively, thereby has greatly improved the extensibility of data center.TVMPP has only considered Internet Transmission flow factor.Some other traditional virtual machine Deployment Algorithm comprises constraint programming, vanning statistics, integer programming, genetic algorithm etc.These methods make data center can use to greatest extent server resource, by close utilization rate very low or and idle server reach the object of saving server energy consumption.Visible, although data center is energy-conservation to have dropped into a large amount of researchers and has studied, the pretty good method of a large amount of performances has also been proposed, but need to provide a kind of from the corresponding Deployment Algorithm of multifactor combination angle research, thereby reach the solution that simultaneously reduces server end energy consumption and network energy consumption double goal, to reduce total energy consumption in most of data center network.
Summary of the invention
In order to overcome the defect of above-mentioned prior art, the invention provides a kind of dynamic energy-saving resource scheduling system and method for distributed cloud computing platform.This system and method is in server end constraints with carry under the dependence constraint multiple condition between the virtual machine of multi-level application program, minimise data central site network transmission quantity and the server wasting of resources, thus network energy consumption and server energy consumption reduced.
In order to realize foregoing invention object, the present invention takes following technical scheme:
A dynamic energy-saving resource scheduling system for distributed cloud computing platform, described system comprises the Cloud Server node under described distributed cloud computing platform, its improvements are: described Cloud Server node connects respectively global controller and monitoring system;
Described global controller is connected respectively modeling analysis module with described monitoring system; Described monitoring system comprises energy consumption monitoring module, flow monitoring module and monitoring resource module; On described Cloud Server node, move Node Controller.
Further, described Cloud Server node comprises each physical server and virtual machine thereof under described distributed cloud computing platform.
Further, described global controller is to realize in the network-wide basis of described distributed cloud computing platform the central controlled module of server energy consumption and network equipment energy consumption in deploying virtual machine process;
Described Node Controller is the module that realizes the resource distribution of virtual machine on physical server.
Further, described energy consumption monitoring module is the monitoring electricity consumption data of described physical server and the module of apparatus cools power information;
Described flow monitoring module is the module of data conversion transmission information, the communication flows data between physical server and the communication flows data between virtual machine of the network equipment in monitoring territory;
The CPU that described monitoring resource module is Statistical Physics server, internal memory, memory space occupied information, virtual machine CPU, internal memory, memory space occupied information, and the module of network equipment CPU, memory information.
Further, the MBM of described modeling analysis module is that operation analysis engine carries out energy consumption, performance, migration modeling, obtains the module of the Optimal Parameters of cloud computing platform performance, energy consumption, migration management;
The analysis module of described modeling analysis module is for realizing the overall situation or the local energy consumption minimized module of described distributed cloud computing platform according to described Optimal Parameters.
A method that is applied to the dynamic energy-saving resource scheduling system of distributed cloud computing platform as claimed in claim 1, its improvements are: described dynamic energy-saving resource regulating method comprises the following steps:
I, energy consumption monitoring module, flow monitoring module and monitoring resource module are obtained respectively consumption information, flow information and the resource information of physical server;
The MBM operation analysis engine of II, modeling analysis module carries out modeling;
The analysis module of III, modeling analysis module carries out the overall situation or local energy consumption minimized;
Domain controller under IV, global controller is realized dynamic energy-saving scheduling of resource.
Further, in described step I, described consumption information comprises: the electricity consumption data of physical server and apparatus cools power information;
Described flow information comprises: the data conversion transmission information of the network equipment, the communication flows data between physical server and the communication flows data between virtual machine in territory;
Described resource information comprises: the CPU of described physical server, internal memory, memory space occupied information, virtual machine CPU, internal memory, memory space occupied information on described physical server, and network equipment CPU, memory information.
Further, described Step II comprises: use the MBM of described modeling analysis module to carry out modeling to energy consumption, performance, migration, obtain performance, the energy consumption of described cloud computing platform, the Optimal Parameters of migration management; Set up and reduce physical server energy consumption and network equipment energy consumption multiple target Optimized model;
Described Step II I comprises: the multiple target Optimized model that the analysis module of modeling analysis module is set up according to described MBM issues the configuration parameter of performance management, managing power consumption, migration management, cross-domain management, realizes the whole network of described distributed cloud computing platform or local energy consumption minimized.
Further, described step IV comprises: the domain controller under described global controller is implemented deploying virtual machine and migration by Cloud Server node, completes dynamic energy-saving scheduling of resource.
Further, the algorithm of described dynamic energy-saving scheduling of resource comprises the following steps:
I, for n virtual machine, need to dispose, dependence matrix between described virtual machine is D, set the node that a described n virtual machine is the dependence graph G between described virtual machine, described dependence matrix D is weighting limit, obtains the dependence graph G between described virtual machine;
The resource requirement of setting resource surplus threshold value iteration, each virtual machine is that the resource of a d dimensional vector, described physical server comprises that the resource requirement of elementary cell, described elementary cell is a d dimensional vector;
II, described figure G is divided into a separate k subset, a described n virtual machine is divided into the k group varying in size, obtain described dependence graph G minimal weight cut k-cut;
III, given resource elementary cell, divide k group by physical server with the integral multiple of elementary cell;
IV, the grouping of matching virtual machine and physical server grouping, until matching result meets resource surplus, be less than or equal to resource surplus threshold value iteration and finish, obtain deploying virtual machine or migration results and met the requirement that minimizes cloud computing platform transmission volume and server resource surplus simultaneously, realize various dimensions and reduce cloud computing platform energy consumption.
Compared with prior art, beneficial effect of the present invention is:
1, system and method for the present invention, for cross-domain characteristic distributions and the synthesis energy saving scheduling of resource demand of distributed cloud computing platform, is realized the effect that simultaneously reduces server and network energy consumption.
2, system and method for the present invention is by introducing the concept in territory, resource management mechanism based on territory is provided, efficiently solve conventional method and carry out with physical cluster granularity the inflexible problem of scheduling that scheduling of resource exists, realized in cluster and across logic reorganization, dynamic assignment and many granularities of cluster resource and having dispatched.Territory has following three kinds with the corresponding relation of cluster: the one, and cluster and territory corresponding one by one, now deteriorates to the scheduling of conventional cluster granularity; The 2nd, can in a larger cluster, carry out the division in territory, realize the fine granularity of resource and distribute and scheduling; The 3rd, can realize scheduling of resource across cluster, logically a plurality of collection are divided into a territory, realize cross-domain unified resource management and scheduling.
3, system and method for the present invention is dispatched by the mode of many class resources scheduling controlling mechanism, the monitoring of energy consumption/resource/flow multidimensional, analysis modeling engine formation closed-loop system is completed to resource dynamic, avoided conventional method to lack the drawback that effective feedback mechanism exists, provide scheduling of resource-> Real-time Feedback-> model parameter of high efficient and flexible to adjust--the scheduling of resource system of the dynamic tuning of > scheduling of resource.
4, the method for the present invention complete information that monitoring provides based on various dimensions (energy consumption, resources occupation rate, communication flows), propose to reduce the Multiobjective Scheduling model of network equipment energy consumption and server energy consumption simultaneously, realized the dispatching algorithm based on k-cut.Compare traditional single goal Energy-saving Building modeling method, avoided the problem of the single and weak effect of simple reduction physical server energy consumption or network equipment energy consumption means; Compare traditional multiple target Energy-saving Building modeling method, the feedback mechanism forming by above-mentioned closed-loop system, capable of dynamic is adjusted relevant parameter, has effectively avoided that multiple target iterative algorithm convergence rate is slow, parameter arranges the problems such as improper.
Accompanying drawing explanation
Fig. 1 is the schematic diagram of dynamic energy-saving resource scheduling system of the present invention;
Fig. 2 is dynamic energy-saving scheduling of resource course of work flow chart of the present invention;
Fig. 3 is dynamic energy-saving resource regulating method flow chart of the present invention.
Embodiment
Below in conjunction with accompanying drawing, the invention will be further described.
Cloud computing platform Huo Yun data center energy consumption mainly comprises server energy consumption and network energy consumption, the key that reduces server energy consumption is to reduce to the full extent the waste of physical machine resource in deploying virtual machine process, use few physical machine of trying one's best carries all virtual machines, maximizes the utilance of physical machine; The key that reduces network energy consumption is to reduce to the full extent switch-spanning data transfer throughput in cloud computing platform by application perception deploying virtual machine mechanism, saves bandwidth resources and switch energy consumption; Therefore the invention provides a kind of dynamic energy-saving resource scheduling system and dispatching method thereof of distributed cloud computing platform, this system and method is for realizing in server end constraints and carrying under the dependence constraint multiple condition between the virtual machine of multi-level application program, minimise data central site network transmission quantity and the server wasting of resources, thereby reduce network energy consumption and server energy consumption, finally realize the target that various dimensions reduce cloud computing platform energy consumption.Above-mentioned various dimensions refer to network equipment energy consumption dimension and server energy consumption dimension.
In method of the present invention, the object that reduces network energy consumption and server energy consumption based on need realizes two class resource dynamic mappings under large-scale distributed cloud computing environment, comprising: one, from application program, load to the mapping of resource requirement, for reducing network energy consumption; Two, the mapping from virtual machine to physical machine, for reducing server energy consumption.
For supporting this two kinds of mappings, system is divided into global controller, domain controller and Node Controller three-level structure.Wherein, global controller is operated in the resource pool level of whole cloud computing platform, and domain controller is operated in management domain level, and Node Controller is operated in virtual machine level.
On each physical server node, move Node Controller, for monitoring, the application resource of virtual machine and carrying thereof takies, energy consumption etc.Domain controller and global controller program have been responsible for respectively deploying virtual machine and the corresponding resource of universe or whole cloud platform and have been distributed.
In addition, the monitoring system of cloud computing platform detects following system information: the energy consumption on each server and the network equipment, resource service condition, the communication flows being mutually related between application program, and form relevant information and send to collective analysis storehouse, analysis and modeling module utilizes the information that cloud platform monitoring system is collected to analyze and set up model to server and network equipment energy consumption, and final result is sent to global controller helps it to complete energy-efficient deployment decision-making.
As shown in Figure 1, Fig. 1 is the schematic diagram of dynamic energy-saving resource scheduling system of the present invention; System of the present invention comprises: the Cloud Server node under distributed cloud computing platform, global controller, Node Controller, modeling analysis module.
State Cloud Server node and connect respectively global controller and Node Controller; Described global controller is connected respectively modeling analysis module with described Node Controller; Described Node Controller comprises energy consumption monitoring module, flow monitoring module and monitoring resource module.
The test machine that Cloud Server node under above-mentioned distributed cloud computing platform is cloud computing platform, carries out dynamic energy-saving resource scheduling system for above-mentioned test machine.Cloud Server node comprises each physical server node under described distributed cloud computing platform, specifically comprises the virtual machine on each physical server and each physical server.
Described global controller connects each domain controller, and described domain controller connects the Node Controller of the above-mentioned Cloud Server node of part in a plurality of physical clusters or in whole physical cluster or in certain physical cluster.
Global controller: the centralized control that realizes in the network-wide basis of described distributed cloud computing platform server energy consumption and network equipment energy consumption in deploying virtual machine process.
Node Controller: run on physical server the module of distributing for realizing the resource of virtual machine on physical server.
Energy consumption monitoring module: for monitoring the electricity consumption data of described physical server and the module of apparatus cools power information.
Flow monitoring module: for monitoring the data conversion transmission information of the network equipment in territory, communication flows data between physical server and the module of the communication flows data between virtual machine.
Monitoring resource module: for CPU, internal memory, the memory space occupied information of Statistical Physics server, virtual machine CPU, internal memory, memory space occupied information, and the module of the information such as network equipment CPU, internal memory.
Modeling analysis module: operation analysis engine carries out energy consumption, performance, migration modeling, obtain the Optimal Parameters of cloud computing platform performance, energy consumption, migration management, the overall situation or the part of according to described Optimal Parameters, realizing described distributed cloud computing platform are energy consumption minimized.
Modeling analysis module comprises MBM and analysis module, wherein, operation analysis engine carries out energy consumption, performance, migration modeling, the Optimal Parameters that obtains cloud computing platform performance, energy consumption, migration management is realized by MBM, realizes the overall or local energy consumption minimized of described distributed cloud computing platform realized by analysis module according to described Optimal Parameters.
As shown in Figure 2, Fig. 2 is dynamic energy-saving scheduling of resource course of work flow chart of the present invention; The present invention also provides a kind of method that is applied to the dynamic energy-saving resource scheduling system of distributed cloud computing platform, and this dynamic energy-saving resource regulating method comprises the following steps:
I, energy consumption monitoring module, flow monitoring module and monitoring resource module are obtained respectively consumption information, flow information and the resource information of physical server;
The MBM operation analysis engine of II, modeling analysis module carries out modeling;
The analysis module of III, modeling analysis module is realized the overall situation or local energy consumption minimized;
Domain controller under IV, global controller is realized dynamic energy-saving scheduling of resource.
In step I, described consumption information comprises: the electricity consumption data of physical server and apparatus cools power information;
Described flow information comprises: the data conversion transmission information of the network equipment, the communication flows data between physical server and the communication flows data between virtual machine in territory;
Described resource information comprises: the CPU of described physical server, internal memory, memory space occupied information, virtual machine CPU, internal memory, memory space occupied information on described physical server, and network equipment CPU, internal memory.
Step II comprises: use described analysis engine to carry out modeling to energy consumption, performance, migration, obtain performance, the energy consumption of described cloud computing platform, the Optimal Parameters of migration management; Set up and reduce physical server energy consumption and network equipment energy consumption multiple target Optimized model.
Step II I comprises: the multiple target Optimized model that global controller is set up according to modeling analysis module issues the configuration parameter of performance management, managing power consumption, migration management, cross-domain management, realizes the whole network of described distributed cloud computing platform or local energy consumption minimized.
Step IV comprises: the domain controller under described global controller is implemented deploying virtual machine and migration by Cloud Server node, completes dynamic energy-saving scheduling of resource.
As shown in Figure 3, Fig. 3 is dynamic energy-saving resource regulating method flow chart of the present invention; The method of above-mentioned dynamic energy-saving scheduling of resource, the method is realized by analysis module, and method comprises the following steps:
I, suppose need to dispose for n virtual machine, the dependence matrix between virtual machine is D, and the cell value that D matrix is is 0 or 1, if two virtual machines exist intercommunication mutually, is 1, there is no correspondence, is 0;
The resource requirement of each virtual machine is a d dimensional vector, and every one dimension representative of this vector operates in a certain class resource of the application requests on virtual machine, comprises CPU, internal memory or memory space etc.
Correspondingly, the resource that physical server possesses is divided into a plurality of elementary cells, the resource of each unit number according to the resource of virtual machine request, determine, resource elementary cell is similarly a d dimensional vector.
Meanwhile, given resource surplus threshold parameter.
For the virtual machine frequently of communicating by letter is assigned on the lower-cost groove of a group communication as far as possible, to reduce the impact that between virtual machine, communication causes bottom topological structure, reach the object that reduces communications cost and reduce network equipment energy consumption; By n virtual machine being regarded as to the node of dependence graph, and dependence matrix D is regarded the weighting limit of dependence graph as, thereby obtains the dependence graph G between virtual machine;
II, above-mentioned dependence graph G is divided into a separate k subset, a described n virtual machine is divided into the k group varying in size, obtain described dependence graph G minimal weight cut k-cut;
III, given resource elementary cell, divide k group by physical server with the integral multiple of elementary cell;
IV, the grouping of matching virtual machine and physical server grouping, until matching result meets resource surplus, be less than or equal to resource surplus threshold value iteration and finish, obtain deploying virtual machine or migration results and met the requirement that minimizes cloud computing platform transmission volume and server resource surplus simultaneously, realize various dimensions (comprising network equipment energy consumption dimension and server energy consumption dimension) and reduce cloud computing platform energy consumption.
One specific embodiment is provided, supposes that the virtual machine of having disposed is according to the enforcement of above-mentioned dynamic resource scheduling mechanism, this embodiment is increment embodiment, and scheme is as follows:
Step 1: distributed cloud platform virtual machine rapid deployment module (described rapid deployment module is the intrinsic module of cloud platform) is to global controller application virtual machine creating and deployment, and the parameter that global controller is submitted to according to user is screened satisfactory physical server node from Global Resource Picture;
Step 2: global controller filters out after available Cloud Server node set, further build the dependence graph G (needing the common participation of monitoring modular and modeling analysis module) of virtual machine, finally determine territory and the carrying server node of deploying virtual machine;
Step 3: Node Controller completes the map bindings process of resources of virtual machine and physical server resource according to the resource scheduling algorithm shown in Fig. 3, and mapping relations are fed back to domain controller and global controller, completes preliminary deployment.
Step 4: the resource using information of the resource using information of the virtual machine that the continuous collection of Node Controller has been disposed, communication flows information, this physical server node, and feed back to analysis modeling module, analysis modeling module is adjusted iterative parameter (as resource surplus threshold value, physical server node resource CPU/ internal memory/memory space threshold value, energy consumption upper limit threshold etc.) according to real time information, and feed back to global controller, by global controller, determine whether to need further to carry out the operations such as virtual machine (vm) migration.
Finally should be noted that: above embodiment is only in order to illustrate that technical scheme of the present invention is not intended to limit, although the present invention is had been described in detail with reference to above-described embodiment, those of ordinary skill in the field are to be understood that: still can modify or be equal to replacement the specific embodiment of the present invention, and do not depart from any modification of spirit and scope of the invention or be equal to replacement, it all should be encompassed in the middle of claim scope of the present invention.
Claims (10)
1. a dynamic energy-saving resource scheduling system for distributed cloud computing platform, described system comprises the Cloud Server node under described distributed cloud computing platform, it is characterized in that: described Cloud Server node connects respectively global controller and monitoring system;
Described global controller is connected respectively modeling analysis module with described monitoring system; Described monitoring system comprises energy consumption monitoring module, flow monitoring module and monitoring resource module; On described Cloud Server node, move Node Controller.
2. dynamic energy-saving resource scheduling system as claimed in claim 1, is characterized in that: described Cloud Server node comprises each physical server and virtual machine thereof under described distributed cloud computing platform.
3. dynamic energy-saving resource scheduling system as claimed in claim 1, is characterized in that: described global controller is to realize in the network-wide basis of described distributed cloud computing platform the central controlled module of server energy consumption and network equipment energy consumption in deploying virtual machine process;
Described Node Controller is the module that realizes the resource distribution of virtual machine on physical server.
4. dynamic energy-saving resource scheduling system as claimed in claim 1, is characterized in that: described energy consumption monitoring module is the monitoring electricity consumption data of described physical server and the module of apparatus cools power information;
Described flow monitoring module is the module of data conversion transmission information, the communication flows data between physical server and the communication flows data between virtual machine of the network equipment in monitoring territory;
The CPU that described monitoring resource module is Statistical Physics server, internal memory, memory space occupied information, virtual machine CPU, internal memory, memory space occupied information, and the module of network equipment CPU, memory information.
5. dynamic energy-saving resource scheduling system as claimed in claim 1, it is characterized in that: the MBM of described modeling analysis module is that operation analysis engine carries out energy consumption, performance, migration modeling, obtain the module of the Optimal Parameters of cloud computing platform performance, energy consumption, migration management;
The analysis module of described modeling analysis module is for realizing the overall situation or the local energy consumption minimized module of described distributed cloud computing platform according to described Optimal Parameters.
6. a method that is applied to the dynamic energy-saving resource scheduling system of distributed cloud computing platform as claimed in claim 1, is characterized in that: described dynamic energy-saving resource regulating method comprises the following steps:
I, energy consumption monitoring module, flow monitoring module and monitoring resource module are obtained respectively consumption information, flow information and the resource information of physical server;
The MBM operation analysis engine of II, modeling analysis module carries out modeling;
The analysis module of III, modeling analysis module carries out the overall situation or local energy consumption minimized;
Domain controller under IV, global controller is realized dynamic energy-saving scheduling of resource.
7. method as claimed in claim 6, is characterized in that: in described step I, described consumption information comprises: the electricity consumption data of physical server and apparatus cools power information;
Described flow information comprises: the data conversion transmission information of the network equipment, the communication flows data between physical server and the communication flows data between virtual machine in territory;
Described resource information comprises: the CPU of described physical server, internal memory, memory space occupied information, virtual machine CPU, internal memory, memory space occupied information on described physical server, and network equipment CPU, memory information.
8. method as claimed in claim 6, it is characterized in that: described Step II comprises: use the MBM of described modeling analysis module to carry out modeling to energy consumption, performance, migration, obtain performance, the energy consumption of described cloud computing platform, the Optimal Parameters of migration management; Set up and reduce physical server energy consumption and network equipment energy consumption multiple target Optimized model;
Described Step II I comprises: the multiple target Optimized model that the analysis module of modeling analysis module is set up according to described MBM issues the configuration parameter of performance management, managing power consumption, migration management, cross-domain management, realizes the whole network of described distributed cloud computing platform or local energy consumption minimized.
9. method as claimed in claim 6, is characterized in that: described step IV comprises: the domain controller under described global controller is implemented deploying virtual machine and migration by Cloud Server node, completes dynamic energy-saving scheduling of resource.
10. method as claimed in claim 9, is characterized in that: the algorithm of described dynamic energy-saving scheduling of resource comprises the following steps:
I, for n virtual machine, need to dispose, dependence matrix between described virtual machine is D, set the node that a described n virtual machine is the dependence graph G between described virtual machine, described dependence matrix D is weighting limit, obtains the dependence graph G between described virtual machine;
The resource requirement of setting resource surplus threshold value iteration, each virtual machine is that the resource of a d dimensional vector, described physical server comprises that the resource requirement of elementary cell, described elementary cell is a d dimensional vector;
II, described figure G is divided into a separate k subset, a described n virtual machine is divided into the k group varying in size, obtain described dependence graph G minimal weight cut k-cut;
III, given resource elementary cell, divide k group by physical server with the integral multiple of elementary cell;
IV, the grouping of matching virtual machine and physical server grouping, until matching result meets resource surplus, be less than or equal to resource surplus threshold value iteration and finish, obtain deploying virtual machine or migration results and met the requirement that minimizes cloud computing platform transmission volume and server resource surplus simultaneously, realize various dimensions and reduce cloud computing platform energy consumption.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410301191.5A CN104052820A (en) | 2014-06-27 | 2014-06-27 | Dynamic energy-saving resource scheduling system and method for distributed cloud computing platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410301191.5A CN104052820A (en) | 2014-06-27 | 2014-06-27 | Dynamic energy-saving resource scheduling system and method for distributed cloud computing platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104052820A true CN104052820A (en) | 2014-09-17 |
Family
ID=51505171
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410301191.5A Pending CN104052820A (en) | 2014-06-27 | 2014-06-27 | Dynamic energy-saving resource scheduling system and method for distributed cloud computing platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104052820A (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103412919A (en) * | 2013-08-09 | 2013-11-27 | 杭州华为数字技术有限公司 | Dispatching method and device for distributed file system |
| CN104572251A (en) * | 2015-01-30 | 2015-04-29 | 中国联合网络通信集团有限公司 | Virtual machine deploying method and device |
| CN104951427A (en) * | 2015-06-30 | 2015-09-30 | 深圳清华大学研究院 | Distributed computation framework with matrix as center |
| CN105357199A (en) * | 2015-11-09 | 2016-02-24 | 南京邮电大学 | Cloud computing cognitive resource management system and method |
| CN105471759A (en) * | 2016-01-11 | 2016-04-06 | 北京百度网讯科技有限公司 | Network traffic scheduling method and apparatus for data centers |
| CN107193362A (en) * | 2017-05-19 | 2017-09-22 | 成都鼎智汇科技有限公司 | One kind enhancing cloud computing environment energy saver |
| CN108134848A (en) * | 2018-02-07 | 2018-06-08 | 北京航空航天大学 | A kind of SOA system resource optimization methods based on graph theory K- segmentations |
| CN109547239A (en) * | 2018-11-14 | 2019-03-29 | 赵显涛 | Strange land cloud data center management system based on three-layer network framework |
| CN110120979A (en) * | 2019-05-20 | 2019-08-13 | 华为技术有限公司 | A kind of dispatching method, device and relevant device |
| CN110158430A (en) * | 2019-05-08 | 2019-08-23 | 中铁北京工程局集团有限公司 | A kind of automatic plucking laminating machine in bridge concrete face |
| CN111443872A (en) * | 2020-03-26 | 2020-07-24 | 深信服科技股份有限公司 | Distributed storage system construction method, device, equipment and medium |
| CN111562966A (en) * | 2020-04-28 | 2020-08-21 | 北京航空航天大学 | Resource arrangement method of man-machine-object fusion cloud computing platform |
| CN111953788A (en) * | 2020-08-17 | 2020-11-17 | 浪潮云信息技术股份公司 | Large-scale cloud platform |
| CN114125037A (en) * | 2021-10-12 | 2022-03-01 | 能科科技股份有限公司 | Rapid cloud deployment method based on collaboration platform |
| CN114301987A (en) * | 2022-03-07 | 2022-04-08 | 天津市城市规划设计研究总院有限公司 | Dynamic scheduling method and system for virtualized network resources |
| CN116541261A (en) * | 2023-07-06 | 2023-08-04 | 成都睿的欧科技有限公司 | Resource management method and system based on cloud resource monitoring |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102759984A (en) * | 2012-06-13 | 2012-10-31 | 上海交通大学 | Power supply and performance management system for virtualization server cluster |
| CN103685562A (en) * | 2013-12-31 | 2014-03-26 | 湖南师范大学 | Cloud computing system and resource and energy efficiency management method thereof |
-
2014
- 2014-06-27 CN CN201410301191.5A patent/CN104052820A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102759984A (en) * | 2012-06-13 | 2012-10-31 | 上海交通大学 | Power supply and performance management system for virtualization server cluster |
| CN103685562A (en) * | 2013-12-31 | 2014-03-26 | 湖南师范大学 | Cloud computing system and resource and energy efficiency management method thereof |
Non-Patent Citations (1)
| Title |
|---|
| 黄道超: "《智慧云网络动态资源适配关键技术研究》", 《中国博士学位论文全文数据库信息科技辑》 * |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103412919B (en) * | 2013-08-09 | 2016-12-28 | 杭州华为数字技术有限公司 | The dispatching method of a kind of distributed file system and equipment |
| CN103412919A (en) * | 2013-08-09 | 2013-11-27 | 杭州华为数字技术有限公司 | Dispatching method and device for distributed file system |
| CN104572251A (en) * | 2015-01-30 | 2015-04-29 | 中国联合网络通信集团有限公司 | Virtual machine deploying method and device |
| CN104572251B (en) * | 2015-01-30 | 2018-01-26 | 中国联合网络通信集团有限公司 | Virtual machine deployment method and device |
| CN104951427A (en) * | 2015-06-30 | 2015-09-30 | 深圳清华大学研究院 | Distributed computation framework with matrix as center |
| CN105357199B (en) * | 2015-11-09 | 2018-05-18 | 南京邮电大学 | A kind of cloud computing cognitive resources manage system and method |
| CN105357199A (en) * | 2015-11-09 | 2016-02-24 | 南京邮电大学 | Cloud computing cognitive resource management system and method |
| CN105471759A (en) * | 2016-01-11 | 2016-04-06 | 北京百度网讯科技有限公司 | Network traffic scheduling method and apparatus for data centers |
| CN105471759B (en) * | 2016-01-11 | 2018-06-01 | 北京百度网讯科技有限公司 | The network traffics dispatching method and device of data center |
| CN107193362A (en) * | 2017-05-19 | 2017-09-22 | 成都鼎智汇科技有限公司 | One kind enhancing cloud computing environment energy saver |
| CN107193362B (en) * | 2017-05-19 | 2020-02-07 | 湖南三湘银行股份有限公司 | Energy-saving device for enhancing cloud computing environment |
| CN108134848B (en) * | 2018-02-07 | 2020-08-04 | 北京航空航天大学 | SOA system resource optimization method based on graph theory K-segmentation |
| CN108134848A (en) * | 2018-02-07 | 2018-06-08 | 北京航空航天大学 | A kind of SOA system resource optimization methods based on graph theory K- segmentations |
| CN109547239A (en) * | 2018-11-14 | 2019-03-29 | 赵显涛 | Strange land cloud data center management system based on three-layer network framework |
| CN110158430A (en) * | 2019-05-08 | 2019-08-23 | 中铁北京工程局集团有限公司 | A kind of automatic plucking laminating machine in bridge concrete face |
| CN110158430B (en) * | 2019-05-08 | 2021-03-02 | 中铁北京工程局集团有限公司 | Automatic napping laminating machine for bridge concrete surface |
| CN110120979A (en) * | 2019-05-20 | 2019-08-13 | 华为技术有限公司 | A kind of dispatching method, device and relevant device |
| CN110120979B (en) * | 2019-05-20 | 2023-03-10 | 华为云计算技术有限公司 | Scheduling method, device and related equipment |
| CN111443872A (en) * | 2020-03-26 | 2020-07-24 | 深信服科技股份有限公司 | Distributed storage system construction method, device, equipment and medium |
| CN111562966A (en) * | 2020-04-28 | 2020-08-21 | 北京航空航天大学 | Resource arrangement method of man-machine-object fusion cloud computing platform |
| CN111562966B (en) * | 2020-04-28 | 2022-09-02 | 北京航空航天大学 | Resource arrangement method of man-machine-object fusion cloud computing platform |
| CN111953788A (en) * | 2020-08-17 | 2020-11-17 | 浪潮云信息技术股份公司 | Large-scale cloud platform |
| CN114125037A (en) * | 2021-10-12 | 2022-03-01 | 能科科技股份有限公司 | Rapid cloud deployment method based on collaboration platform |
| CN114301987A (en) * | 2022-03-07 | 2022-04-08 | 天津市城市规划设计研究总院有限公司 | Dynamic scheduling method and system for virtualized network resources |
| CN116541261A (en) * | 2023-07-06 | 2023-08-04 | 成都睿的欧科技有限公司 | Resource management method and system based on cloud resource monitoring |
| CN116541261B (en) * | 2023-07-06 | 2023-09-05 | 成都睿的欧科技有限公司 | Resource management method and system based on cloud resource monitoring |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104052820A (en) | Dynamic energy-saving resource scheduling system and method for distributed cloud computing platform | |
| CN110416998B (en) | Regional complex distribution network scheduling control management system based on virtual power plant | |
| WO2023201916A1 (en) | Distributed flexible resource aggregation control apparatus and control method | |
| CN103384272B (en) | A kind of cloud service distributive data center system and load dispatching method thereof | |
| CN103150003B (en) | The power-saving control method of information handling system, this information handling system and device | |
| CN107579518B (en) | Power system environment economic load dispatching method and apparatus based on MHBA | |
| Phan et al. | Evolutionary multiobjective optimization for green clouds | |
| CN113675853B (en) | Energy internet-oriented electricity consumption information acquisition system | |
| CN108429265B (en) | Demand response regulation and control method and device | |
| CN102508714A (en) | Green-computer-based virtual machine scheduling method for cloud computing | |
| CN110365052A (en) | Microgrid energy-storage system state consistency control method based on power optimization scheduling | |
| CN107221965B (en) | Day-ahead plan calculation method based on distributed design | |
| CN107423133B (en) | Data network load distribution method among data centers for reducing power grid loss | |
| CN109617099B (en) | Virtual energy storage coordination control system and method thereof | |
| CN105446816A (en) | Heterogeneous platform oriented energy consumption optimization scheduling method | |
| Yang et al. | An optimal battery allocation model for battery swapping station of electric vehicles | |
| CN108808737A (en) | Promote the active distribution network Optimization Scheduling of renewable distributed generation resource consumption | |
| Zhou et al. | Spatial–temporal energy management of base stations in cellular networks | |
| CN109284336B (en) | Geographically distributed data center system and scheduling method thereof | |
| CN102325054A (en) | Self-adaptive adjusting method for hierarchy management of distributed type calculation management platform cluster | |
| Zhao et al. | Workload and energy management of geo-distributed datacenters considering demand response programs | |
| CN117526311A (en) | Electric automobile cluster control method based on genetic algorithm | |
| CN103176850A (en) | Electric system network cluster task allocation method based on load balancing | |
| Xu et al. | A virtual data center deployment model based on the green cloud computing | |
| CN102523300A (en) | Data-intensive cloud storage model facing intelligent power grid |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140917 |