CN102662754A - Multi-field supportable virtual machine dispatching device and multi-field supportable virtual machine dispatching method - Google Patents
Multi-field supportable virtual machine dispatching device and multi-field supportable virtual machine dispatching method Download PDFInfo
- Publication number
- CN102662754A CN102662754A CN2012101191647A CN201210119164A CN102662754A CN 102662754 A CN102662754 A CN 102662754A CN 2012101191647 A CN2012101191647 A CN 2012101191647A CN 201210119164 A CN201210119164 A CN 201210119164A CN 102662754 A CN102662754 A CN 102662754A
- Authority
- CN
- China
- Prior art keywords
- virtual machine
- scheduling
- cluster
- virtual
- information
- 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
Images
Landscapes
- Stored Programmes (AREA)
Abstract
The invention discloses a multi-field supportable virtual machine dispatching device and a multi-field supportable virtual machine dispatching method. The multi-field supportable virtual machine dispatching device comprises a virtual machine cluster information central access module, a virtual machine dispatching module and a virtual machine dispatching information central output module, wherein the virtual machine cluster information central access module is used for receiving input cluster information, the virtual machine dispatching module is used for analyzing cluster information and generating dispatching policy required by fields, and the virtual machine dispatching information central output module is used for centrally outputting policy information. The virtual machine cluster information central access module receives processing requests of a virtualization cluster system and preprocesses the same into a uniform input information format. The virtual dispatching module classifies states of the virtualization cluster system according to input information, generates corresponding dispatching policy for processing the requests according to classifications and outputs dispatching commands through the information central output module. Dispatching virtualization clusters under multiple fields can be realized well, and the purpose of dispatching under different fields is achieved. By modular design, uniform interfaces are used among modules, transparency in design of dispatching algorithm is improved, and application range is wide.
Description
Technical field
The invention belongs to virtual Clustering field, be specifically related to a kind of scheduling virtual machine apparatus and method of supporting under many scenes.
Background technology
In recent years, increasingly mature along with Intel Virtualization Technology, the field of application constantly increases, and the scale of virtual cluster is also increasing.In the middle of the application of large-scale virtual cluster, because the demand difference of different business field and application scenarios causes the purpose of virtual colony dispatching there are differences.Simply, the scheduling virtual machine device of simplification can not satisfy the dispatching requirement of industry to virtual cluster.Need the monitoring continuously and the analysis of resource utilizations such as CPU, internal memory, the network bandwidth, disk through physical machine in the cluster; According to the different application scenes requirement; The appropriate virtual machine in the cluster moves scheduling, reaches the purpose of reasonable distribution resource.
In the application of reality, a lot of common scenes are arranged, different scene has different purpose and demand, for example:
The energy-saving and emission-reduction scene is an important use scene, and it is in the primary demand that guarantees resources of virtual machine; Consider to open the physical machine of minimum number,, integrate the virtual machine on the different physical machine through to the part virtual machine (vm) migration; The soar physical machine of partial redundance is closed.Through based on the scheduling under the energy-saving and emission-reduction scene, reduce the energy consumption cost of physical machine cluster, to reach the purpose of energy-saving and emission-reduction.In the middle of this scene, need take all factors into consideration service disconnection, resource consumption in the transition process, avoid closing repeatedly and opening physical machine in the short time simultaneously.
The load balancing scene is according to certain rule, and rational each item resource of dynamic assignment and balance physical machine makes in the cluster resource utilization of each physical machine suitable.After adopting load balance scheduling, the too high or too low situation of the average rate of profit of the relative cluster of some physical machine resource utilization can not appear in the cluster.
Focus is removed scene, in the middle of this scene, pays close attention to the physical machine of load too high.Through constantly obtaining and analyze the loading condition of every physical machine in the cluster,, confirm the physical machine of load too high according to the threshold value of reality.Through moving virtual machine on this physical machine on other physical machine or the physical machine newly opened, remove with the focus that reaches cluster.
Maximum resource utilization factor scene, the resource utilization of the whole cluster of concern in the middle of this scene.In order to reach maximization to resource utilizations such as cluster CPU, internal memory, the network bandwidth, disks, need be to the resource operating position of monitoring and analyzing virtual machine.Through certain migration and deployment strategy, reasonably distribute the virtual machine on each physical machine, guarantee virtual cluster the better response time is arranged in to reach the peak use rate of cluster resource.
Current existing dispatching system is just developed to the certain applications scene, to satisfy demand separately.Though the particular problem of a certain moment of solution that this way can be temporary transient, lacking can expansion, dirigibility and self adaptive.When new scene occurring is to expand effectively, and what new dispatching algorithm can not be very fast is integrated in the middle of the dispatching system, also has a large amount of overlapping development simultaneously, is unfavorable for the long term growth of system of virtual cluster.
Summary of the invention
The object of the present invention is to provide a kind of scheduling virtual machine apparatus and method of supporting under many scenes, can be according to the different scene requirement, through monitoring and analyzing virtual cluster resource user mode, the scheduling strategy of virtual machine under the suitable different scenes is provided.
A kind of scheduling virtual machine device of supporting many scenes is applied in the virtual cluster virtual machine moved scheduling, it is characterized in that, comprises successively connecting:
Cluster virtual machine information is unified access modules, is used to receive the processing request of system of virtual cluster, passes to subsequent module through generation input information after the pre-service and analyzes and handle;
The scheduling virtual machine module is used to receive said scheduling virtual machine information and unifies access modules through pretreated input information, generates scheduling strategy;
Scheduling virtual machine information is unified output module, is used to receive the scheduling strategy of said scheduling virtual machine module, presses arranged standard format organization and handles scheduling strategy, and scheduling and the decision instruction of handling the back generation are sent to system of virtual cluster.
Further, described scheduling virtual machine module comprises:
Cluster integrality judging unit is used to judge the state of said system of virtual cluster;
The virtual machine scheduling policy selected cell according to the state of said system of virtual cluster, is selected concrete scheduling strategy unit;
The scheduling strategy unit generates concrete scheduling strategy, and outputs to scheduling virtual machine information and unify output module.
Further, described scheduling strategy unit comprises the initial placement unit, and focus is removed the unit, and Load Balance Unit is newly opened the physical machine unit, dynamically integral unit.Call corresponding scheduling strategy unit according to different scenes and analyze and handle, generate different scheduling strategies.
Wherein, described processing request comprises the physical machine of said cluster virtual machine and the resource operating position information of virtual machine, virtual machine dependence information, information of forecasting.Cluster virtual machine information is unified access modules these information is carried out pre-service, is transferred to the scheduling virtual machine module with the form of XML file, has comprised the current time load information and some algorithmic dispatching configuration informations of whole all virtual machines of cluster.
Wherein, unified calling interface and action command are deferred in described each scheduling strategy unit.The scheduling virtual machine module provides unified interface between each scheduling strategy unit, defers to unified action command form, and each unit can be selected according to the different scene needs dynamically.Because the present invention has adopted scene drive mechanism and unified calling interface and action command is provided, and can carry out hot plug to each scheduling strategy unit in the scheduler module easily.
The present invention also provides a kind of dispatching method of virtual machine of supporting many scenes, is applied in the virtual cluster virtual machine moved scheduling, comprises step:
Step 2, receive pretreated input information, with system of virtual cluster state classification;
Step 3, generate corresponding scheduling strategy according to the state of current system of virtual cluster;
Step 4, with scheduling strategy change into the scheduling and decision instruction give system of virtual cluster.
Wherein, said system of virtual cluster state is divided into overload HOT, idle COLD, three kinds of states of normal load WARM, and HOT representes the entire system load too high, and COLD representes that the entire system load is lower, and WARM representes that the entire system load is moderate.
Further, said step 3 comprises:
Current system of virtual cluster state is the HOT state, generates the scheduling strategy of newly opening physical machine;
When the system of virtual cluster load is in WARM or COLD state,, then generates and place virtual machine scheduling policy if the information of input has the request of initial placement virtual machine; If there is not the request of initial placement, then check whether there is the focus physical machine in the current cluster, if there is focus, generate migration part virtual machine scheduling policy; If there is not the focus physical machine, and the system of virtual cluster load is in COLD, then generates and dynamically integrates scheduling strategy; If there is not the focus physical machine, and the system of virtual cluster load is in WARM, then generates the load balance scheduling strategy.
Wherein, described processing request comprises the physical machine of said cluster virtual machine and the resource operating position information of virtual machine, virtual machine dependence information, information of forecasting.
Wherein, input information and scheduling strategy are with the form transmission of XML file.Each intermodule shields the complicacy of input/output information through the transmission of unified XML formatted file in the scheduling virtual machine device.
The present invention has standard and complete virtual colony dispatching architectural framework, uses the present invention can support the scheduling of the virtual cluster under many scenes preferably, arrives the scheduling purpose under the different scenes.It is that the invention and the design of virtual colony dispatching under many scenes supplied good method and standard that the present invention simultaneously has adaptivity, controllability and Scalable Performance preferably, has improved the transparency of dispatching algorithm design, has wide range of applications.
Description of drawings
Fig. 1: support the scheduling virtual machine device under many scenes for the present invention is a kind of;
Fig. 2: be the service logic synoptic diagram of scheduler module of the present invention.
Embodiment
Below in conjunction with accompanying drawing and embodiment technical scheme of the present invention is explained further details, following examples do not constitute qualification of the present invention.
As shown in Figure 1; Scheduling virtual machine device of the present invention comprises that the cluster virtual machine information that receives the cluster information input unifies access modules, analyzes cluster information and requires to produce the scheduling virtual machine module of scheduling decision with scene and will dispatch the scheduling virtual machine information that unification is exported with decision information and unify output module.
System of virtual cluster can send the request of processing to scheduling virtual machine device of the present invention as required at any time, comprises initial placement, focus releasing, load balancing, energy-saving and emission-reduction, startup scheduling etc.It is that cluster provides information to unify input interface that cluster virtual machine information is unified access modules; Receive the processing request that system of virtual cluster is sent; The request of processing comprises resource operating position (CPU, internal memory, IO, the network) information of physical machine and virtual machine; Virtual machine dependence (static configuration, the performance impact) information that business model is confirmed, the information of forecasting that the data learning model obtains.Cluster virtual machine information passes to the scheduling virtual machine module through generation input information after the pre-service and analyzes and handle after unifying access modules reception processing request.
The scheduling virtual machine module is the core of whole dispatching system, and the scheduling virtual machine module receives input information, and the integrality of cluster is discerned.Scheduler module is selected different scheduling strategy unit according to the condition intelligent of being discerned, and forms different scheduling decisions.The scheduling virtual machine module comprises:
Cluster integrality judging unit is used to judge the state of said cluster virtual machine system;
The virtual machine scheduling policy selected cell according to the state of said system of virtual cluster, is selected concrete scheduling strategy unit;
The scheduling strategy unit generates concrete scheduling strategy, and outputs to scheduling virtual machine information and unify output module, and the scheduling strategy unit comprises the initial placement unit, and focus is removed the unit, and Load Balance Unit is newly opened the physical machine unit, dynamically integral unit.
Scheduling virtual machine module sink virtual machine schedule information is unified the data of access modules input; Take all factors into consideration and reduce migration cost (service disconnection, resource consumption etc.), avoid resource fragmentation, reduce the migration number of times; Prevent the migration concussion; Avoid the factors such as load balancing of physical machine power-on and power-off repeatedly, cluster virtual machine, call corresponding scheduling strategy unit according to different scenes and analyze and handle, and the scheduling strategy after will handling passes to scheduling virtual machine information and unifies output module.Scheduling and decision instruction comprise: migration instruction, following electricity are specified, instruction powers on.Wherein the migration instruction comprises source host, destination host, virtual machine to be migrated, and the instruction that powers on comprises the platform number that need newly open physical machine, every electric down instruction comprises a pent physical machine sign of needs.The scheduling virtual machine module provides unified interface between each scheduling strategy unit, defers to unified action command form, and each unit can be selected according to the different scene needs dynamically.Because the present invention has adopted scene drive mechanism and unified calling interface and action command is provided, and can carry out hot plug to each scheduling strategy unit in the scheduler module easily.
Scheduling virtual machine information is unified the schedule information of output module sink virtual machine scheduler module, presses arranged standard format organization and handles scheduling strategy.The scheduling of the generation after it will be handled and decision instruction send to the cluster virtual machine system.
Unify access modules through scheduling virtual machine information and can be transformed into the input information that can supply the scheduling virtual machine module to use to the status information of the physical machine of complicacy and virtual machine.Unify the elementary instruction that output module can supply scheduler module information transition module the physical cluster scheduling through scheduling virtual machine information.Through with each module package, adopt the design of stack, the complicacy of the input/output information of shielding.Each intermodule is through the transmission of unified XML formatted file in the present embodiment scheduling virtual machine device.
As shown in Figure 2, scheduling virtual machine module logic function of the present invention comprises: the cluster state analysis, and business logic processing, the result of decision generates.Concrete dispatching method is following:
The scheduling virtual machine module receives the cluster load data that system of virtual cluster sends over, and load data has comprised the current time load information and some algorithmic dispatching configuration informations of whole all virtual machines of cluster with the form transmission of XML file;
The cluster condition judgment module of scheduling virtual machine module is analyzed input information; According to data cases, learn current entire system situation, confirm that current system is in corresponding state HOT, WARM, (overall load is too high for COLD; Overall load is moderate, and overall load is low excessively).
Like Fig. 2, current cluster load too high, physical cluster is in the HOT state, with directly sending the decision-making of newly opening physical machine, and finishes current scheduling, can carry out the next round decision-making in case of necessity, alleviates too high load pressure on the existing physical machine through the migration virtual machine.
When the cluster load is in WARM or COLD state, can select to carry out remaining service logic.According to the information of input, at first judge the current request that whether has the initial placement virtual machine, then make the decision-making of placing physical machine if exist, the result of decision-making is placed in output.What be worth explanation is, can be convenient to use or design voluntarily different placement algorithms according to different demands here, has shielded the complicacy of concrete system.If there is not the request of initial placement, then can select to get into focus and remove the unit, check whether there is the focus physical machine in the current cluster virtual machine; If there is the focus physical machine; Can through dispatching system migration part virtual machine, and can finish this scheduling according to the scene demand.If there is not the focus physical machine, then get into different service logic dynamically integration or load balancing according to system state and scene demand.
If the cluster load is in COLD, then can directly carry out the dynamic integral unit of virtual machine, reach energy-saving and emission-reduction.
If the cluster load is in WARM, then can directly carry out Load Balance Unit, in Load Balance Unit, can select different equalization algorithms according to the input of entire system state.
At last, scheduling virtual machine information is unified output module and can be supplied cluster processed instruction sequence to give system of virtual cluster result of decision parsing module, accomplishes whole scheduling process.
Because the virtual machine number in the system and the variation of its load condition load between COLD, WARM, the HOT whole cluster and change.According to the different scene demand, can reach dbjective state through multiple scheduling means.The action of system call decision-making can be divided into: open physical machine, close physical machine, close virtual machine, open virtual machine, move virtual machine.Through shielding the complicacy of different scene information inputs and dispatch command, can realize the quick design and the enforcement of each modular algorithm.Demand in the several scenes is interdepending and making up of multiple decision-making scheduling actions, and the thinking that the present invention shifts through module time design clearly, modular approach and load condition realizes supporting the dispatching system of many scenes.
Above embodiment is only in order to technical scheme of the present invention to be described but not limit it; Under the situation that does not deviate from spirit of the present invention and essence thereof; Those of ordinary skill in the art work as can make various corresponding changes and distortion according to the present invention, but these corresponding changes and distortion all should belong to the protection domain of the appended claim of the present invention.
Claims (10)
1. a scheduling virtual machine device of supporting many scenes is applied in the virtual cluster virtual machine moved scheduling, it is characterized in that, comprises connecting successively:
Cluster virtual machine information is unified access modules, is used to receive the processing request of system of virtual cluster, passes to subsequent module through generation input information after the pre-service and analyzes and handle;
The scheduling virtual machine module is used to receive said scheduling virtual machine information and unifies access modules through pretreated input information, generates scheduling strategy;
Scheduling virtual machine information is unified output module, is used to receive the scheduling strategy of said scheduling virtual machine module, presses arranged standard format organization and handles scheduling strategy, and scheduling and the decision instruction of handling the back generation are sent to system of virtual cluster.
2. the scheduling virtual machine device of the many scenes of support according to claim 1 is characterized in that, described scheduling virtual machine module comprises:
Cluster integrality judging unit is used to judge the state of said system of virtual cluster;
The virtual machine scheduling policy selected cell according to the state of said system of virtual cluster, is selected concrete scheduling strategy unit;
The scheduling strategy unit generates concrete scheduling strategy, and outputs to scheduling virtual machine information and unify output module.
3. the scheduling virtual machine device of the many scenes of support according to claim 2 is characterized in that, described scheduling strategy unit comprises the initial placement unit, and focus is removed the unit, and Load Balance Unit is newly opened the physical machine unit, dynamically integral unit.
4. the scheduling virtual machine device of the many scenes of support according to claim 1 and 2 is characterized in that, described processing request comprises the physical machine of said cluster virtual machine and the resource operating position information of virtual machine, virtual machine dependence information, information of forecasting.
5. the scheduling virtual machine device of the many scenes of support according to claim 3 is characterized in that, unified calling interface and action command are deferred in described each scheduling strategy unit.
6. a dispatching method of virtual machine of supporting many scenes is applied in the virtual cluster virtual machine moved scheduling, it is characterized in that, comprises step:
Step 1, reception system of virtual cluster are handled request, carry out pre-service to handling request;
Step 2, receive pretreated input information, with system of virtual cluster state classification;
Step 3, generate corresponding scheduling strategy according to the state of current system of virtual cluster;
Step 4, with scheduling strategy change into the scheduling and decision instruction give system of virtual cluster.
7. the dispatching method of virtual machine of the many scenes of support according to claim 6; It is characterized in that; Said system of virtual cluster state is divided into overload HOT, idle COLD, three kinds of states of normal load WARM; HOT representes the entire system load too high, and COLD representes that the entire system load is lower, and WARM representes that the entire system load is moderate.
8. the dispatching method of virtual machine of the many scenes of support according to claim 7 is characterized in that, said step 3 comprises:
Current system of virtual cluster state is the HOT state, generates the scheduling strategy of newly opening physical machine;
When the system of virtual cluster load is in WARM or COLD state,, then generates and place virtual machine scheduling policy if the information of input has the request of initial placement virtual machine; If there is not the request of initial placement, then check whether there is the focus physical machine in the current cluster, if there is focus, generate migration part virtual machine scheduling policy; If there is not the focus physical machine, and the system of virtual cluster load is in COLD, then generates and dynamically integrates scheduling strategy; If there is not the focus physical machine, and the system of virtual cluster load is in WARM, then generates the load balance scheduling strategy.
9. according to claim 6, the dispatching method of virtual machine of the many scenes of the described support of 7,8 arbitrary claims; It is characterized in that; Described processing request comprises the physical machine of said cluster virtual machine and the resource operating position information of virtual machine, virtual machine dependence information, information of forecasting.
10. according to claim 6, the dispatching method of virtual machine of the many scenes of the described support of 7,8 arbitrary claims is characterized in that, input information and scheduling strategy are with the form transmission of XML file.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012101191647A CN102662754A (en) | 2012-04-20 | 2012-04-20 | Multi-field supportable virtual machine dispatching device and multi-field supportable virtual machine dispatching method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2012101191647A CN102662754A (en) | 2012-04-20 | 2012-04-20 | Multi-field supportable virtual machine dispatching device and multi-field supportable virtual machine dispatching method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102662754A true CN102662754A (en) | 2012-09-12 |
Family
ID=46772253
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2012101191647A Pending CN102662754A (en) | 2012-04-20 | 2012-04-20 | Multi-field supportable virtual machine dispatching device and multi-field supportable virtual machine dispatching method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102662754A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103377281A (en) * | 2012-04-23 | 2013-10-30 | 国际商业机器公司 | Method and system for remediating resource overload in over-committed computing environment |
| CN103605578A (en) * | 2013-12-09 | 2014-02-26 | 武汉理工大学 | Load balance scheduling method based on virtual machine migration |
| CN103677958A (en) * | 2013-12-13 | 2014-03-26 | 华为技术有限公司 | Virtualization cluster resource scheduling method and device |
| CN103970587A (en) * | 2014-04-21 | 2014-08-06 | 华为技术有限公司 | Resource scheduling method, device and system |
| CN104239123A (en) * | 2014-09-05 | 2014-12-24 | 北方工业大学 | Campus-cloud-platform-oriented virtual machine management dispatching method and system |
| CN104461789A (en) * | 2014-12-30 | 2015-03-25 | 华夏银行股份有限公司 | Control method and system for achieving host and backup switching of disaster backup system |
| CN104484222A (en) * | 2014-12-31 | 2015-04-01 | 北京天云融创软件技术有限公司 | Virtual machine dispatching method based on hybrid genetic algorithm |
| CN105335211A (en) * | 2015-10-21 | 2016-02-17 | 同济大学 | System and method for dispatching FPGA (Field Programmable Gate Array) accelerator based on Xen virtualized cluster |
| CN106648871A (en) * | 2016-12-28 | 2017-05-10 | 北京奇艺世纪科技有限公司 | Resource management method and system |
| CN107169129A (en) * | 2017-06-06 | 2017-09-15 | 山东浪潮通软信息科技有限公司 | A kind of dispatching method and device |
| CN108667859A (en) * | 2017-03-27 | 2018-10-16 | 中兴通讯股份有限公司 | A kind of method and device for realizing scheduling of resource |
| CN108833528A (en) * | 2018-06-11 | 2018-11-16 | 郑州云海信息技术有限公司 | A kind of cloud platform colony dispatching method and apparatus |
| CN110825512A (en) * | 2019-11-08 | 2020-02-21 | 东南数字经济发展研究院 | Universal distributed system design and C + + implementation thereof |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100115513A1 (en) * | 2008-10-30 | 2010-05-06 | Hitachi, Ltd. | Virtual machine control method and virtual machine system |
| CN102262567A (en) * | 2010-05-24 | 2011-11-30 | 中兴通讯股份有限公司 | Virtual machine scheduling decision system, platform and method |
-
2012
- 2012-04-20 CN CN2012101191647A patent/CN102662754A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100115513A1 (en) * | 2008-10-30 | 2010-05-06 | Hitachi, Ltd. | Virtual machine control method and virtual machine system |
| CN102262567A (en) * | 2010-05-24 | 2011-11-30 | 中兴通讯股份有限公司 | Virtual machine scheduling decision system, platform and method |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103377281B (en) * | 2012-04-23 | 2016-08-03 | 国际商业机器公司 | The method and system of the overload in the computing environment repairing excessively promise |
| CN103377281A (en) * | 2012-04-23 | 2013-10-30 | 国际商业机器公司 | Method and system for remediating resource overload in over-committed computing environment |
| CN103605578A (en) * | 2013-12-09 | 2014-02-26 | 武汉理工大学 | Load balance scheduling method based on virtual machine migration |
| CN103605578B (en) * | 2013-12-09 | 2017-03-08 | 武汉理工大学 | Load equilibration scheduling method based on virtual machine (vm) migration |
| CN103677958A (en) * | 2013-12-13 | 2014-03-26 | 华为技术有限公司 | Virtualization cluster resource scheduling method and device |
| CN103677958B (en) * | 2013-12-13 | 2017-06-20 | 华为技术有限公司 | The resource regulating method and device of a kind of virtual cluster |
| CN103970587A (en) * | 2014-04-21 | 2014-08-06 | 华为技术有限公司 | Resource scheduling method, device and system |
| CN104239123A (en) * | 2014-09-05 | 2014-12-24 | 北方工业大学 | Campus-cloud-platform-oriented virtual machine management dispatching method and system |
| CN104239123B (en) * | 2014-09-05 | 2019-02-15 | 北方工业大学 | Virtual Machine Manager dispatching method and system towards campus cloud platform |
| CN104461789A (en) * | 2014-12-30 | 2015-03-25 | 华夏银行股份有限公司 | Control method and system for achieving host and backup switching of disaster backup system |
| CN104484222A (en) * | 2014-12-31 | 2015-04-01 | 北京天云融创软件技术有限公司 | Virtual machine dispatching method based on hybrid genetic algorithm |
| CN105335211A (en) * | 2015-10-21 | 2016-02-17 | 同济大学 | System and method for dispatching FPGA (Field Programmable Gate Array) accelerator based on Xen virtualized cluster |
| CN105335211B (en) * | 2015-10-21 | 2018-10-26 | 同济大学 | A kind of FPGA accelerators scheduling system and method based on Xen virtual clusters |
| CN106648871A (en) * | 2016-12-28 | 2017-05-10 | 北京奇艺世纪科技有限公司 | Resource management method and system |
| CN108667859A (en) * | 2017-03-27 | 2018-10-16 | 中兴通讯股份有限公司 | A kind of method and device for realizing scheduling of resource |
| CN107169129A (en) * | 2017-06-06 | 2017-09-15 | 山东浪潮通软信息科技有限公司 | A kind of dispatching method and device |
| CN107169129B (en) * | 2017-06-06 | 2020-07-14 | 浪潮通用软件有限公司 | Scheduling method and device |
| CN108833528A (en) * | 2018-06-11 | 2018-11-16 | 郑州云海信息技术有限公司 | A kind of cloud platform colony dispatching method and apparatus |
| CN110825512A (en) * | 2019-11-08 | 2020-02-21 | 东南数字经济发展研究院 | Universal distributed system design and C + + implementation thereof |
| CN110825512B (en) * | 2019-11-08 | 2024-05-24 | 东南数字经济发展研究院 | Universal distributed system design and C++ implementation thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102662754A (en) | Multi-field supportable virtual machine dispatching device and multi-field supportable virtual machine dispatching method | |
| Chien et al. | Dynamic resource prediction and allocation in C-RAN with edge artificial intelligence | |
| US20160248266A1 (en) | Heterogeneous Battery Cell Charging | |
| CN103885831B (en) | The system of selection of virtual machine host machine and device | |
| CN102707995B (en) | Service scheduling method and device based on cloud computing environments | |
| CN103076870B (en) | Scheduling and dynamic resource allocation method are merged in the application that in data center, energy consumption drives | |
| US20190114206A1 (en) | System and method for providing a performance based packet scheduler | |
| CN103188277A (en) | Load energy consumption management system, method and server | |
| CN102262567A (en) | Virtual machine scheduling decision system, platform and method | |
| CN102611723A (en) | Method for building high-performance computing application service based on virtualization technology | |
| Azizi et al. | A priority-based service placement policy for fog-cloud computing systems | |
| Garg et al. | SDN-NFV-Aided edge-cloud interplay for 5G-envisioned energy internet ecosystem | |
| CN102043676B (en) | Visualized data centre dispatching method and system | |
| CN102455932A (en) | Serial execution method, device and system for task instances | |
| CN103927244A (en) | Plug-in scheduling process monitoring method implemented based on dynamic proxy | |
| Jahangard et al. | Taxonomy of green cloud computing techniques with environment quality improvement considering: a survey | |
| Sadatdiynov et al. | An intelligent hybrid method: Multi-objective optimization for MEC-enabled devices of IoE | |
| RU2609076C2 (en) | Method and system for smart control over distribution of resources in cloud computing environments | |
| Zhang et al. | An energy-aware task scheduling algorithm for a heterogeneous data center | |
| CN102104487A (en) | Message processing method and equipment | |
| CN106357676A (en) | Method for optimizing overhead of cloud service resource | |
| CN106027685A (en) | Peak access method based on cloud computation system | |
| Azizi et al. | A QoS-aware service placement algorithm for fog-cloud computing environments | |
| CN113157383A (en) | Method for dynamically adjusting super-proportion in OpenStack environment | |
| CN105468451A (en) | Job scheduling system of computer cluster on the basis of high-throughput sequencing data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20120912 |