CN102012835A - Virtual central processing unit (CPU) scheduling method capable of supporting software real-time application - Google Patents
Virtual central processing unit (CPU) scheduling method capable of supporting software real-time application Download PDFInfo
- Publication number
- CN102012835A CN102012835A CN2010106128994A CN201010612899A CN102012835A CN 102012835 A CN102012835 A CN 102012835A CN 2010106128994 A CN2010106128994 A CN 2010106128994A CN 201010612899 A CN201010612899 A CN 201010612899A CN 102012835 A CN102012835 A CN 102012835A
- Authority
- CN
- China
- Prior art keywords
- real
- time
- virtual machine
- vcpu
- cpu
- 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.)
- Granted
Links
Images
Landscapes
- Stored Programmes (AREA)
Abstract
The invention discloses a virtual CPU (VCPU) scheduling method capable of supporting software real-time application in a multinuclear virtual environment, which uses a virtual machine real-time engine module and a VCPU operation queue real-time scheduling module. The virtual machine real-time engine module consists of a real-time automatic sensing analysis sub-module and a real-time manual configuration sub-module, wherein the real-time automatic sensing analysis sub-module judges the real-time performance of the virtual machine according to the acquired data on the operation conditions of the virtual machine; and the real-time manual configuration sub-module completes manual setting for meeting the real-time requirements of specific virtual machines. The VCPU operation queue real-time scheduling module consists of a CPU operation queue state analysis sub-module, a VCPU operation queue real-time policy sub-module and a VCPU operation queue optimization sub-module, wherein the CPU operation queue state analysis sub-module acquires the binding information of a VCPU operation queue according to a real-time state; the VCPU operation queue real-time policy sub-module performs VCPU operation queue regulation according to a given policy; and the VCPU operation queue optimization sub-module responses for performing the optimization operation of the VCPU operation queue.
Description
Technical field
The present invention relates to the support method of the technical field of virtualization under the multinuclear hardware platform in the computer science, the particularly soft real-time application of virtual machine.
Background technology
Virtual is the abstract technology of a kind of computer system, by the virtual machine logical layer supervisory computer resource on the physical platform, and provides virtual resource to thereon client layer of operation.Client OS can pass through the standard interface access resources, but interface is not but directly communicated by letter with physical resource; But carry out the management of real resources by virtualization layer, make the resource of limited quantity can be multiplexing between a plurality of passenger planes family.The virtualization layer that runs directly on the hardware platform is commonly referred to as virtual machine monitor (Virtual Machine Monitor, VMM or hypervisor), and it is responsible for safeguarding abstract, guarantees the correctness of resource multiplex simultaneously; The client layer main body is an operating system completely, is called Client OS.Virtual machine is the combination of real machine and virtualization software (Virtualization Software)." virtual " comprises two processes: from the mapping of virtual resource to real resources; The instruction of virtual machine and the desired function of system call are realized by actual instruction and system call.Therefore, virtual machine monitor can promote the efficient utilization of resource, and the isolation of maintain customer machine operation system, provides great dirigibility and maneuverability for system design simultaneously.The development that is widely used as Intel Virtualization Technology of multinuclear hardware platform technology provides opportunities and challenges, and representative is the SVM technology of VT and the AMD of Intel.Just for these reasons, Intel Virtualization Technology is more and more significantly outstanding, and range of application is extensive further.At present, Intel Virtualization Technology is used in enterprise-oriented system, service provider, family's desktop system, mobile device and production system or the like aspect.The Intel Virtualization Technology product of current main-stream comprises VMware and Xen, KVM etc., and they have certain contribution at aspects such as Server Consolidation, resource optimization, energy controls, and progressively have been applied in fields such as cloud computing and high-performance calculation.Intel Virtualization Technology still has the wide prospect of utilizing in the Virtual Cluster field simultaneously.
Soft real-time system is meant system under the heavy load situation, and permission is missed the situation in time limit and can not caused too big harm.It has certain requirement to the response time, but is not strictness very, such as interactive stream media services such as video conference system and VOIP services.At present, also there is the demand that is deployed on the multinuclear virtual platform in soft real-time application service, still, because Intel Virtualization Technology effect aspect the support of soft real-time application is also very undesirable, still needs the further feasible scheme of research.With Intel Virtualization Technology Xen is example, justice and consistance are all paid attention to and emphasized to its CPU scheduling and I/O dispatching algorithm to all virtual machines that move on it, do not consider the concrete time requirement of institute's operation task in virtual machine and the virtual machine, it is from start to finish to all virtual machines and virtual cpu (Virtual CPU thereof, VCPU) all put on an equal footing, carry out different VCPU successively in turn according to basic time-sharing method.Typical soft real-time application is compared with the intensive program of CPU with common I/O is intensive, and the regular hour requirement is arranged, and has such characteristics of demand: I/O processing and wants in time, and CPU calculates and wants synchronous.Though soft real-time application does not need too much CPU computational resource, need the frequent data transmission of carrying out, carry out alternately with the user.The tolerance of program response time is vital for user's physical experience, only finish operation with subscriber-related I/O in the short time, distribute CPU to finish data processing timely, make the data computation of CPU and I/O operational coordination synchronous, just can reach desirable live effect.
Summary of the invention
The present invention has designed the VCPU dispatching method that a kind of multinuclear virtualized environment supports that soft real-time is used, realized the effective support of virtual machine manager to soft real-time task, by the time of the performed task of virtual machine and the perception of characteristics are judged its real-time, and the VCPU formation with real-time is dispatched according to certain strategy, shorten break period and the task response-time of real-time VCPU, support soft real-time application aims thereby reach.
Specifically comprise:
1. virtual machine real-time engine
The present invention is provided with a virtual machine real-time engine in virtual machine monitor, be used to reflect the real-time feature of virtual machine.Virtual machine real-time engine comprises the automatic perception analysis of real-time and two submodules of real-time manual configuration.Wherein, the automatic perception analysis of real-time is automatic perception and the analysis to the demand of virtual machine real-time, it is periodically to the ruuning situation and the data of each virtual machine are added up and calculated for the previous period, according to certain condition judgment go out each virtual machine the real-time that should have, then the real-time engine is configured; Real-time manual configuration submodule is by the real-time requirement of approach such as the management interface configuration particular virtual machine of system platform keeper by virtual machine monitor, finishes final configuration to virtual machine real-time engine by virtual machine monitor again.Virtual machine real-time engine has also been safeguarded a virtual machine real-time mark sheet and a mark sheet update mark, item in the virtual machine real-time mark sheet has marked the real-time feature of corresponding virtual machine, and the mark sheet update mark is only set when the real-time mark sheet changes.
2.VCPU operation queue real-time scheduling
VCPU operation queue real-time scheduler module is used to realize the scheduling support of virtual machine monitor to real-time VCPU.It is to be made of CPU operation queue state analysis, VCPU operation queue real time strategies and three submodules of VCPU operation queue optimization.And each CPU also to a CPU real-time state should be arranged, has comprised the status information of its VCPU operation queue.CPU operation queue state analysis submodule only is set the back in the mark sheet update mark and starts, and analyzes the real-time status information of the operation queue of CPU; VCPU operation queue real time strategies submodule is based on the result that the CPU operation queue is analyzed, and carries out the distribution adjustment of VCPU according to set strategy; It is after each scheduling operation is finished that the VCPU operation queue is optimized submodule, carries out the Optimizing operation of VCPU operation queue, guarantees that promptly all real-time VCPU are in the override position of ad eundem subqueue, and keeps the relative priority order constant.
At virtualized environment to aspect the soft real-time scheduling, also relevant at present ripe research and application achievements, concrete innovative point of the present invention is:
(1) the present invention is not intended for the dispatching algorithm and the realization of any specific Intel Virtualization Technology, and the polytype multi-core platform is had general adaptability;
(2) the present invention has taked the decentralized processing that the real-time VCPU of different virtual machine distributes on CPU, has avoided the execution conflict of real-time VCPU;
(3) the present invention has realized the dynamic binding of VCPU on CPU, has reduced the cpu cache that causes because of too much VCPU migration and has lost efficacy;
Description of drawings
Fig. 1 multinuclear virtualized environment is supported the VCPU dispatching method architecture of soft real-time application down
Real-time perception of Fig. 2 virtual machine and real-time manual configuration method flow diagram
The operation queue strategy execution process flow diagram of Fig. 3 CPU
Fig. 4 real-time VCPU operation queue Optimizing operation synoptic diagram
Embodiment
As shown in Figure 1, support the VCPU dispatching method of soft real-time application to constitute under the multinuclear virtual environment by virtual machine real-time engine and VCPU operation queue real-time scheduling two big modules.Wherein, virtual machine real-time engine modules is the input of VCPU operation queue real-time module, is the foundation as CPU operation queue state analysis and the execution of VCPU operation queue real time strategies.
Virtual machine real-time engine modules comprises the automatic perception analysis of real-time and two submodules of real-time manual configuration.Two submodules are finished the concrete configuration to virtual machine real-time engine jointly.Virtual machine real-time engine has also been safeguarded a virtual machine real-time mark sheet and a mark sheet update mark, and each virtual machine has marked the real-time feature of this virtual machine corresponding to one in the virtual machine real-time mark sheet; This real-time feature is free, perception real-time and three kinds of states of manual configuration real-time, is initially set to sky.Wherein, real-time is characterized as sky, represents this virtual machine not have the real-time feature; Real-time is characterized as the perception real-time, and representative draws this virtual machine by the automatic perception analysis submodule of real-time and has the real-time feature; Real-time is characterized as the manual configuration real-time, and representative draws this virtual machine by real-time manual configuration submodule and has the real-time feature.Virtual machine with real-time feature is meant that the feature of this virtual machine in the real-time mark sheet be not for empty; VCPU with virtual machine of real-time feature is called real-time VCPU, and the mark sheet update mark is set when the real-time mark sheet changes, in order to start the function of VCPU real-time module.
The automatic perception analysis submodule of real-time is that the requirement virtual machine monitor periodically obtains the real-time running data about each virtual machine, detect and add up each virtual machine, draw the real-time of virtual machine according to following Rule of judgment in for the previous period unit interval I/O number of operations and VCPU single scheduled for executing time:
Particular virtual machine unit interval I/O number of operations 〉=virtual machine average unit interval I/O number of operations * α (i)
Particular virtual machine VCPU single scheduled for executing time≤the average single scheduled for executing of virtual machine VCPU time * β (ii)
(the α initial value is 2 more than the mean value α of system times for being higher than at unit interval I/O number of operations, and can dynamically change and adjust), simultaneously the VCPU single scheduled for executing time is lower than the mean value β of system doubly following (β is initialized as 0.5, and virtual machine meeting dynamic change adjustment), promptly satisfy under (i) and the situation (ii), the capital is judged as the virtual machine that real-time requires automatically, thereby the virtual machine real-time mark sheet to virtual machine real-time engine is configured, and the real-time mark sheet respective items assignment that is about to this virtual machine is the perception real-time.
Real-time manual configuration submodule requires the control interface of system platform keeper by virtual machine monitor, finishes configuration to the real-time demand of particular virtual machine by virtual machine real-time engine, and upgrades virtual machine real-time mark sheet.
Above process as shown in Figure 2, automatic perception analysis parameter alpha of real-time wherein and β require to have certain adaptability, the dynamic adjustment that virtual machine quantity that the real-time feature is arranged that the each perception of needs basis is judged and manual configuration situation are carried out parameter, when the virtual machine quantity of perception real-time state is too much, then heighten α automatically and reasonably, turn down β, then turn down α on the contrary, heighten β, can not surpass the CPU number with the virtual machine quantity of guaranteeing to be configured to have real-time feature.
VCPU operation queue real-time scheduler module comprises CPU operation queue state analysis, VCPU operation queue real time strategies and three submodules of VCPU operation queue optimization.It is provided with a CPU real-time state for each CPU, and this real-time state has comprised the status information of its VCPU operation queue.
CPU operation queue state analysis submodule starts when the real-time update mark is set up, eliminate the real-time update mark after the startup immediately, inquire about and analyze the real-time state of each CPU then, obtain the specifying information that the binding of CPU operation queue has real-time feature virtual machine.That is, whether CPU is bound has real-time feature virtual machine, be specially which virtual machine, and this virtual machine is arranged in the real-time VCPU quantity of this CPU operation queue.
VCPU operation queue real time strategies comprises:
(1) real-time VCPU adjusts the override position that its order that is scheduled is in ad eundem VCPU when being waken up by states such as obstruction or off-lines.
(2) all the real-time VCPU with virtual machine of real-time feature dynamically bind on the appropriate C PU, and can not moved away by load balance mechanism, also do not conflict mutually with other real-time VCPU.
(3) principle that manual configuration real-time virtual machine has precedence over automatic perception analysis is observed in the configuration with virtual machine of real-time feature.
(4) guarantee to have the number of the no more than actual physics CPU of quantity of the virtual machine of real-time feature.
(5) any VCPU migration operation of the original dispatching algorithm of virtual machine monitor is invalid to the real-time VCPU of dynamic binding, and promptly scheduler can not be moved to real-time VCPU on the operation queue of other CPU.
(6) virtual machine of cancellation real-time feature in time is distributed to its VCPU on other CPU by former dynamic binding CPU.
The realization of VCPU operation queue real time strategies as shown in Figure 3, it analyzes status information that submodule obtains and above-mentioned real time strategies according to the CPU operation queue, carries out the adjustment of VCPU operation queue according to following two kinds of situations respectively:
(1) the real-time state of this CPU shows that it has dynamically been bound a virtual machine with real-time feature, and mark real-time VCPU quantity, the adjustment of VCPU operation queue strategy then is responsible for checking the integrality of real-time state, and carries out necessary maintenance concerns.The integrality of CPU real-time state is meant that the operation queue of this CPU comprises and only include the whole VCPU with real-time feature virtual machine that dynamically bound.Maintenance concerns then is under the ungratified situation of integrality, the necessity maintenance work of carrying out comprises that migration has destroyed other real-time VCPU of real-time state integrity, interpolation belongs to bound and has other VCPU of real-time feature virtual machine in this formation;
(2) CPU is not dynamically bound any virtual machine with real-time feature, VCPU operation queue real time strategies submodule then needs traversal to check this VCPU operation queue, if find to have real-time VCPU, whether then inquire about the affiliated virtual machine of this VCPU has been bound on other CPU, if then move out of this real-time VCPU, if not, just the virtual machine under this real-time VCPU is dynamically bound on this CPU, and upgrade the real-time state of this CPU, carry out again (1).
It is after each scheduling operation is finished that the VCPU operation queue is optimized submodule, carries out the Optimizing operation of VCPU operation queue, guarantees that promptly all real-time VCPU are in the override position of ad eundem subqueue, and keeps the relative priority order constant.Traditional VCPU scheduler is ranked according to priority, do not consider the real-time of VCUP, the VCPU operation queue is optimized submodule by real-time VCPU is shifted to an earlier date in the position of VCPU operation queue, make real-time VCPU in time to be responded, as shown in Figure 4, as real-time VCPU, VCPU 1.0 on the CPU i and VCPU 1.1, and the position of the VCPU on the CPUj 3.0 in the VCPU operation queue all obtains after the optimization of VCPU operation queue in advance, VCPU 0.1 on the CPU j is not owing to itself just at team's head, therefore adjust.
Claims (7)
1. a multi-core platform is supported the VCPU dispatching method of soft real-time application under virtualized environment, comprises virtual machine real-time engine and VCPU operation queue real-time scheduling two big modules, it is characterized in that:
(1) virtual machine real-time engine is used to reflect the real-time feature of virtual machine specifically be divided into two submodules of automatic perception analysis of real-time and real-time manual configuration, wherein,
The automatic perception analysis submodule requirement of real-time virtual machine monitor periodically obtains the real-time running data about each virtual machine, detect and add up each virtual machine, draw the real-time demand of virtual machine according to given Rule of judgment in for the previous period unit interval I/O number of operations and VCPU single scheduled for executing time;
Real-time manual configuration submodule is by the real-time requirement of approach such as the management interface configuration particular virtual machine of system platform keeper by virtual machine monitor, finishes final configuration to virtual machine real-time engine by virtual machine monitor again;
In addition, virtual machine real-time engine has also been safeguarded a virtual machine real-time mark sheet and a mark sheet update mark, item in the virtual machine real-time mark sheet has marked the real-time feature of corresponding virtual machine, and update mark is set to mark sheet when the real-time mark sheet changes;
(2) VCPU operation queue real-time scheduler module is used to realize the scheduling support to real-time virtual machine VCPU, it is to be made of CPU operation queue state analysis, VCPU operation queue real time strategies and three submodules of VCPU operation queue optimization, and each CPU is also to there being a CPU real-time state, the status information that has comprised its VCPU operation queue, wherein
CPU operation queue state analysis submodule only is set the back in the mark sheet update mark and starts, and analyzes the real-time state of CPU, obtains the specifying information of real-time virtual machine of the binding of VCPU operation queue;
VCPU operation queue real time strategies submodule is based on the result that the CPU operation queue is analyzed, and carries out the distribution adjustment of VCPU according to set strategy;
It is after each scheduling operation is finished that the VCPU operation queue is optimized submodule, carries out the Optimizing operation of VCPU operation queue, guarantees that promptly all real-time VCPU are in the override position of ad eundem subqueue, and keeps the relative priority order constant.
2. the method for claim 1 is characterized in that, virtual machine of each correspondence in the virtual machine real-time mark sheet, and virtual machine real-time feature comprises sky, perception real-time and three kinds of states of manual configuration real-time, is initially set to sky; Wherein, real-time is characterized as sky, represents this virtual machine not have the real-time feature; Real-time is characterized as the perception real-time, and representative draws this virtual machine by the automatic perception analysis submodule of real-time and has the real-time feature; Real-time is characterized as the manual configuration real-time, and representative draws this virtual machine by real-time manual configuration submodule and has the real-time feature.
3. the method for claim 1, it is characterized in that, Rule of judgment in the automatic perception analysis submodule of real-time is: particular virtual machine unit interval I/O number of operations 〉=virtual machine average unit interval I/O number of operations * α (i) particular virtual machine VCPU single scheduled for executing time≤the average single scheduled for executing of virtual machine VCPU time * β (ii) satisfies condition (i) simultaneously and condition virtual machine (ii) is judged as the real-time virtual machine, α wherein is the coefficient of unit interval I/O number of operations when judging the virtual machine real-time, β then is the coefficient of single scheduled for executing time, α and β can dynamically adjust, and basic span is respectively [2, + ∞) with (0,0.5] half-open intervals, initial value is respectively 2 and 0.5; By adjusting α and β perceptual parameters, guarantee that the number of real-time virtual machine can be more than the number of actual CPU; When autosense real-time virtual machine quantity is too much, then heighten α automatically and reasonably, turn down β, then turn down α on the contrary, heighten β.
4. the method for claim 1 is characterized in that, the specific strategy of VCPU operation queue real time strategies submodule is:
1) real-time VCPU adjusts the override position that its order that is scheduled is in ad eundem VCPU when being waken up by states such as obstruction or off-lines;
2) all the real-time VCPU with virtual machine of real-time feature dynamically bind on the appropriate C PU, and can not moved away by load balance mechanism, also do not conflict mutually with other real-time VCPU;
3) principle that manual configuration real-time virtual machine has precedence over automatic perception analysis is observed in the configuration with virtual machine of real-time feature;
4) guarantee to have the number of the no more than actual physics CPU of quantity of the virtual machine of real-time feature;
5) any VCPU migration operation of the original dispatching algorithm of virtual machine monitor is invalid to the real-time VCPU of dynamic binding, and promptly scheduler can not be moved to real-time VCPU on the operation queue of other CPU;
6) virtual machine of cancellation real-time feature in time is distributed to its VCPU on other CPU by former dynamic binding CPU.
5. the method for claim 1 is characterized in that, whether the realization of VCPU operation queue real time strategies binds the virtual machine with real-time feature according to CPU is divided into following two kinds of situations:
1) the real-time state of this CPU shows that it has dynamically been bound a virtual machine with real-time feature, and mark real-time VCPU quantity, the adjustment of VCPU operation queue strategy then is responsible for checking the integrality of real-time state, and carries out necessary maintenance concerns; The integrality of CPU real-time state is meant that the operation queue of this CPU comprises and only include the whole VCPU with real-time feature virtual machine that dynamically bound; Maintenance concerns then is under the ungratified situation of integrality, the necessity maintenance work of carrying out comprises that migration has destroyed other real-time VCPU of real-time state integrity, interpolation belongs to bound and has other VCPU of real-time feature virtual machine in this formation;
2) CPU is not dynamically bound any virtual machine with real-time feature, VCPU operation queue real time strategies submodule then needs traversal to check this VCPU operation queue, if find to have real-time VCPU, whether then inquire about the affiliated virtual machine of this VCPU has been bound on other CPU, if then move out of this real-time VCPU, if not, just the virtual machine under this real-time VCPU is dynamically bound on this CPU, and upgrade the real-time state of this CPU, carry out again (1).
6. the method for claim 1 is characterized in that, the operating system of virtual machine is Linux.
7. the method for claim 1 is characterized in that, virtual machine monitor requires to support the multicore architecture of SMP.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 201010612899 CN102012835B (en) | 2010-12-22 | 2010-12-22 | Virtual central processing unit (CPU) scheduling method capable of supporting software real-time application |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 201010612899 CN102012835B (en) | 2010-12-22 | 2010-12-22 | Virtual central processing unit (CPU) scheduling method capable of supporting software real-time application |
Publications (2)
Publication Number | Publication Date |
---|---|
CN102012835A true CN102012835A (en) | 2011-04-13 |
CN102012835B CN102012835B (en) | 2013-03-27 |
Family
ID=43843011
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN 201010612899 Expired - Fee Related CN102012835B (en) | 2010-12-22 | 2010-12-22 | Virtual central processing unit (CPU) scheduling method capable of supporting software real-time application |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102012835B (en) |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102521055A (en) * | 2011-12-15 | 2012-06-27 | 广州杰赛科技股份有限公司 | Virtual machine resource allocating method and virtual machine resource allocating system |
WO2012152159A1 (en) * | 2011-11-24 | 2012-11-15 | 中兴通讯股份有限公司 | Scheduling method and device in virtual operating system |
CN102831013A (en) * | 2012-07-19 | 2012-12-19 | 西安交通大学 | VOD (Video On Demand) application resource consumption prediction method based on virtual machine |
CN103440159A (en) * | 2013-08-05 | 2013-12-11 | 华为技术有限公司 | Method and system for scheduling processes |
WO2013189343A2 (en) * | 2013-01-31 | 2013-12-27 | 中兴通讯股份有限公司 | Adaptive virtual machine boot control method and apparatus |
CN103870313A (en) * | 2012-12-17 | 2014-06-18 | ***通信集团公司 | Virtual machine task scheduling method and system |
CN103886033A (en) * | 2014-03-05 | 2014-06-25 | 无锡香象生物科技有限公司 | Intelligent vertical searching device and method for safety industry chain |
CN103927225A (en) * | 2014-04-22 | 2014-07-16 | 浪潮电子信息产业股份有限公司 | Multi-core framework Internet information processing and optimizing method |
CN104123174A (en) * | 2014-08-06 | 2014-10-29 | 华中科技大学 | Dynamic real-time CPU scheduling system under virtualization environment |
CN104160376A (en) * | 2012-03-07 | 2014-11-19 | 微软公司 | Offline provisioning of virtual machines |
CN104765613A (en) * | 2015-04-21 | 2015-07-08 | 华中科技大学 | Optimization method oriented to task parallel programming model under virtualization environment |
CN105279012A (en) * | 2015-11-27 | 2016-01-27 | 成都微讯云通科技有限公司 | Cloud platform job scheduling method |
CN107479945A (en) * | 2017-08-15 | 2017-12-15 | 爱普(福建)科技有限公司 | A kind of resources of virtual machine dispatching method and device |
CN111427669A (en) * | 2020-04-27 | 2020-07-17 | 安谋科技(中国)有限公司 | Method, apparatus, medium, and system for managing virtual machines on computer device |
CN116501447A (en) * | 2023-06-20 | 2023-07-28 | 麒麟软件有限公司 | Xen-based hard real-time implementation system |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090037926A1 (en) * | 2007-08-01 | 2009-02-05 | Peter Dinda | Methods and systems for time-sharing parallel applications with performance isolation and control through performance-targeted feedback-controlled real-time scheduling |
CN101488098A (en) * | 2009-02-13 | 2009-07-22 | 华中科技大学 | Multi-core computing resource management system based on virtual computing technology |
CN101894047A (en) * | 2010-06-24 | 2010-11-24 | 北京航空航天大学 | Kernel virtual machine scheduling policy-based implementation method |
-
2010
- 2010-12-22 CN CN 201010612899 patent/CN102012835B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20090037926A1 (en) * | 2007-08-01 | 2009-02-05 | Peter Dinda | Methods and systems for time-sharing parallel applications with performance isolation and control through performance-targeted feedback-controlled real-time scheduling |
CN101488098A (en) * | 2009-02-13 | 2009-07-22 | 华中科技大学 | Multi-core computing resource management system based on virtual computing technology |
CN101894047A (en) * | 2010-06-24 | 2010-11-24 | 北京航空航天大学 | Kernel virtual machine scheduling policy-based implementation method |
Cited By (28)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2012152159A1 (en) * | 2011-11-24 | 2012-11-15 | 中兴通讯股份有限公司 | Scheduling method and device in virtual operating system |
CN103136045A (en) * | 2011-11-24 | 2013-06-05 | 中兴通讯股份有限公司 | Dispatching method and device of virtualization operating system |
CN102521055A (en) * | 2011-12-15 | 2012-06-27 | 广州杰赛科技股份有限公司 | Virtual machine resource allocating method and virtual machine resource allocating system |
CN104160376B (en) * | 2012-03-07 | 2017-07-28 | 微软技术许可有限责任公司 | The offline preparation of virtual machine |
CN104160376A (en) * | 2012-03-07 | 2014-11-19 | 微软公司 | Offline provisioning of virtual machines |
CN102831013B (en) * | 2012-07-19 | 2014-11-05 | 西安交通大学 | VOD (Video On Demand) application resource consumption prediction method based on virtual machine |
CN102831013A (en) * | 2012-07-19 | 2012-12-19 | 西安交通大学 | VOD (Video On Demand) application resource consumption prediction method based on virtual machine |
CN103870313B (en) * | 2012-12-17 | 2017-02-08 | ***通信集团公司 | Virtual machine task scheduling method and system |
CN103870313A (en) * | 2012-12-17 | 2014-06-18 | ***通信集团公司 | Virtual machine task scheduling method and system |
WO2013189343A2 (en) * | 2013-01-31 | 2013-12-27 | 中兴通讯股份有限公司 | Adaptive virtual machine boot control method and apparatus |
WO2013189343A3 (en) * | 2013-01-31 | 2014-02-20 | 中兴通讯股份有限公司 | Adaptive virtual machine boot control method and apparatus |
CN103440159B (en) * | 2013-08-05 | 2017-04-19 | 华为技术有限公司 | Method and system for scheduling processes |
CN103440159A (en) * | 2013-08-05 | 2013-12-11 | 华为技术有限公司 | Method and system for scheduling processes |
CN103886033A (en) * | 2014-03-05 | 2014-06-25 | 无锡香象生物科技有限公司 | Intelligent vertical searching device and method for safety industry chain |
CN103886033B (en) * | 2014-03-05 | 2017-02-08 | 无锡香象生物科技有限公司 | Intelligent vertical searching device and method for safety industry chain |
CN103927225A (en) * | 2014-04-22 | 2014-07-16 | 浪潮电子信息产业股份有限公司 | Multi-core framework Internet information processing and optimizing method |
CN103927225B (en) * | 2014-04-22 | 2018-04-10 | 浪潮电子信息产业股份有限公司 | A kind of internet information processing optimization method of multi-core framework |
CN104123174A (en) * | 2014-08-06 | 2014-10-29 | 华中科技大学 | Dynamic real-time CPU scheduling system under virtualization environment |
CN104123174B (en) * | 2014-08-06 | 2017-06-16 | 华中科技大学 | Dynamic realtime CPU scheduling system under a kind of virtualized environment |
CN104765613A (en) * | 2015-04-21 | 2015-07-08 | 华中科技大学 | Optimization method oriented to task parallel programming model under virtualization environment |
CN104765613B (en) * | 2015-04-21 | 2017-09-12 | 华中科技大学 | Towards the optimization method of tasks in parallel programming model under a kind of virtualized environment |
CN105279012A (en) * | 2015-11-27 | 2016-01-27 | 成都微讯云通科技有限公司 | Cloud platform job scheduling method |
CN107479945A (en) * | 2017-08-15 | 2017-12-15 | 爱普(福建)科技有限公司 | A kind of resources of virtual machine dispatching method and device |
CN107479945B (en) * | 2017-08-15 | 2021-06-22 | 爱普(福建)科技有限公司 | Virtual machine resource scheduling method and device |
CN111427669A (en) * | 2020-04-27 | 2020-07-17 | 安谋科技(中国)有限公司 | Method, apparatus, medium, and system for managing virtual machines on computer device |
WO2021218935A1 (en) * | 2020-04-27 | 2021-11-04 | 安谋科技(中国)有限公司 | Method and apparatus for managing virtual machine on computer device, and medium and system |
CN116501447A (en) * | 2023-06-20 | 2023-07-28 | 麒麟软件有限公司 | Xen-based hard real-time implementation system |
CN116501447B (en) * | 2023-06-20 | 2023-09-26 | 麒麟软件有限公司 | Xen-based hard real-time implementation system |
Also Published As
Publication number | Publication date |
---|---|
CN102012835B (en) | 2013-03-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102012835B (en) | Virtual central processing unit (CPU) scheduling method capable of supporting software real-time application | |
CN102662763B (en) | Virtual machine resource scheduling method based on service quality | |
Singh et al. | Workflow task clustering for best effort systems with Pegasus | |
Vázquez et al. | On the use of clouds for grid resource provisioning | |
CN103069389B (en) | High-throughput computing method and system in a hybrid computing environment | |
CN102521055B (en) | Virtual machine resource allocating method and virtual machine resource allocating system | |
CN102135903B (en) | Hardware virtualization-based Xen real-time performance improving system and method thereof | |
JPWO2008132924A1 (en) | Virtual computer system and optimization method thereof | |
Cheng et al. | vScale: Automatic and efficient processor scaling for SMP virtual machines | |
CN102253857A (en) | Xen virtual machine scheduling control method in multi-core environment | |
CN104598298A (en) | Virtual machine dispatching algorithm based on task load and current work property of virtual machine | |
Salimi et al. | Advantages, challenges and optimizations of virtual machine scheduling in cloud computing environments | |
US20090133029A1 (en) | Methods and systems for transparent stateful preemption of software system | |
US20120240111A1 (en) | Storage medium storing program for controlling virtual machine, computing machine, and method for controlling virtual machine | |
KR101330609B1 (en) | Method For Scheduling of Mobile Multi-Core Virtualization System To Guarantee Real Time Process | |
Thawari et al. | An efficient data locality driven task scheduling algorithm for cloud computing | |
Shih et al. | Fairness scheduler for virtual machines on heterogonous multi-core platforms | |
US11550625B2 (en) | Utilizing machine learning to concurrently optimize computing resources and licenses in a high-performance computing environment | |
Nemati et al. | Resource sharing among real-time components under multiprocessor clustered scheduling | |
CN103777995A (en) | Synchronization optimized dispatching system and dispatching method based on group dispatching in virtualization environment | |
Zhang et al. | Profit based two-step job scheduling in clouds | |
Gupta | Techniques for efficient high performance computing in the cloud | |
Zhou et al. | Resource allocation in cloud computing based on clustering method | |
Wang et al. | ACS: an effective admission control scheme with deadlock resolutions for workflow scheduling in clouds | |
Groesbrink | Basics of virtual machine migration on heterogeneous architectures for self-optimizing mechatronic systems: Necessary conditions and implementation issues |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130327 Termination date: 20141222 |
|
EXPY | Termination of patent right or utility model |