CN107479944A - Mix the adaptive thermophoresis dispatching method of virutal machine memory and system under cloud mode - Google Patents
Mix the adaptive thermophoresis dispatching method of virutal machine memory and system under cloud mode Download PDFInfo
- Publication number
- CN107479944A CN107479944A CN201710597298.2A CN201710597298A CN107479944A CN 107479944 A CN107479944 A CN 107479944A CN 201710597298 A CN201710597298 A CN 201710597298A CN 107479944 A CN107479944 A CN 107479944A
- Authority
- CN
- China
- Prior art keywords
- copy
- dirty pages
- containing dirty
- current containing
- pages rate
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/505—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the load
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/45583—Memory management, e.g. access or allocation
Abstract
The invention provides a kind of adaptive thermophoresis dispatching method of virutal machine memory and system mixed under cloud mode, including：Structure containing dirty pages rate forecast model carrys out current containing dirty pages rate of the prediction source virutal machine memory after iteration copy；An iteration copy is carried out to source virtual machine internal memory, by the comparison of current containing dirty pages rate and current containing dirty pages rate threshold value, determines that performing continuation iteration copy step still performs shutdown copy and copy on demand；Source virtual machine internal memory is carried out to continue iteration copy, according to the number for continuing iteration copy, or the comparison according to current containing dirty pages rate and current containing dirty pages rate threshold value, decide whether to enter and shut down copy；Wherein, current containing dirty pages rate threshold value is equal to the average value of the current containing dirty pages rate of all applications in source virtual machine internal memory.The present invention can effectively reduce the bulk migration time, and total internal storage data migration amount greatly reduces；Meanwhile for the self-adapting virtual machine scheduling of resource under different application scene.
Description
Technical field
The present invention relates to technical field of data processing, in particular it relates to which the virutal machine memory resource under mixing cloud mode is certainly
Adapt to thermophoresis dispatching method.
Background technology
Current cloud computing resources are widely used in commercial kitchen area, and increasing enterprise and developer are by their service
Including the mixing high in the clouds of private clound and public cloud with application deployment.But due to being limited by resources of virtual machine, and
Load balancing is considered between virtual machine, and enterprise and developer targetedly must carry out thermophoresis to virtual machine according to demand.
Live migration of virtual machine is also referred to as dynamic migration (thermophoresis), be serviced in ensureing virtual machine it is continual under the premise of, will be virtual
Machine resource moves to the process of destination host from source host.In order to which the virtual machine ensured to move on destination host can recover work
Make and service is provided, it is necessary to enough virtual machine informations, including internal memory, CPU, disk, network, I/ are migrated to destination host
The influence of the status informations such as O, wherein memory information to live migration of virtual machine performance is maximum, is the focus of current thermophoresis research,
Three kinds of basic steps of memory copying include：Iteration copy, shut down copy and copy on demand.
As shown in figure 1, most common virutal machine memory thermomigration process mainly has pre-copy (Pre-copy) and rear copy
(Post-copy) two methods, certain two kinds in three kinds of basic steps of memory copying, such as pre-copy elder generation both only be make use of
Copy (Push copy) is iterated, shutdown copy (Stop-and-copy) is carried out after the copy of internal memory iteration several times,
Copy the virtual machine state information on all source hosts to destination host, complete thermophoresis；After copy, first once stopped
Machine copies, and reruns virtual machine in destination host, subsequently into (Pull copy) stage is copied on demand, is directed to
Property from source host copy memory pages, so as to complete thermophoresis.
According to the difference of application type in virtual machine, two methods are each advantageous.For example, answering for memory-intensive
With due to needing substantial amounts of read/write memory, constantly producing substantial amounts of internal memory containing dirty pages face, be copied after being adapted to use in this case
Algorithm；Again for example, the application for not a large amount of memory read-writes, then when being more suitable for using pre-copy algorithm to save internal memory migration
Between.But the adaptive scheduling that existing virutal machine memory duplication technology also neither one can be directed under different application scene is calculated
Method, it is difficult to reach internal memory migration time, migrating data amount and the whole machine balancing in migration applicability.
Time series autoregression model (Autoregressive Model) is with the process for itself doing regression variable, i.e. profit
The linear regression model (LRM) of certain moment stochastic variable after being described with the linear combination of the stochastic variable at early stage at some moment, it is
A kind of common form in time series.AR forecast models are widely used among the prediction to time series, are expanded afterwards
It is used for solving variable prediction and optimization problem to other field.
The content of the invention
For in the prior art the defects of, it is an object of the invention to provide it is a kind of mix cloud mode under virutal machine memory from
Adapt to thermophoresis dispatching method and system.
According to the adaptive thermophoresis dispatching method of virutal machine memory under mixing cloud mode provided by the invention, including：
Containing dirty pages rate forecast model construction step：Structure containing dirty pages rate forecast model carrys out prediction source virutal machine memory to be copied in iteration
Current containing dirty pages rate afterwards；
An iteration copies step：An iteration copy is carried out to source virtual machine internal memory, by current containing dirty pages rate and currently
The comparison of containing dirty pages rate threshold value, determine that performing continuation iteration copy step still performs shutdown copy and copy on demand；
Continue iteration copy step：Source virtual machine internal memory is carried out to continue iteration copy, according to time for continuing iteration copy
Number, or the comparison according to current containing dirty pages rate and current containing dirty pages rate threshold value, decide whether to enter and shut down copy；
Wherein, current containing dirty pages rate threshold value is equal to the average value of the current containing dirty pages rate of all applications in source virtual machine internal memory.
Preferably, in an iteration copies step, the state of current containing dirty pages rate threshold value is exceeded in current containing dirty pages rate
Under, perform and shut down copy and copy on demand, in the state of current containing dirty pages rate is no more than current containing dirty pages rate threshold value, execution continues to change
Generation copy step.
Preferably, in the continuation iteration copy step, in the state of continuing iteration copy number and reaching threshold value, hold
Row shuts down copy.
Preferably, in the continuation iteration copy step, it is less than the state of current containing dirty pages rate threshold value in current containing dirty pages rate
Under, perform and shut down copy.
Preferably, the source virtual machine internal memory of low containing dirty pages rate is preferentially copied in the continuation iteration copy step, is being shut down
Remaining source virtual machine internal memory is disposably copied in copy.
According to a kind of adaptive thermophoresis scheduling system of virutal machine memory mixed under cloud mode provided by the invention, bag
Include：
Containing dirty pages rate forecast model：Current containing dirty pages rate of the prediction source virutal machine memory after iteration copy；
Internal memory migration control module：An iteration copy is carried out to source virtual machine internal memory, by current containing dirty pages rate and currently
The comparison of containing dirty pages rate threshold value, determine to perform and continue iteration copy or perform to shut down copy and copy on demand；Continue to change in execution
For in copy procedure, according to the number for continuing iteration and copying, or the comparison according to current containing dirty pages rate and current containing dirty pages rate threshold value,
Decide whether to enter and shut down copy；
Wherein, current containing dirty pages rate threshold value is equal to the average value of the current containing dirty pages rate of all applications in source virtual machine internal memory.
Preferably, in an iteration copy, in the state of current containing dirty pages rate exceedes current containing dirty pages rate threshold value, institute
State internal memory migration control module to perform shutdown copy and copy on demand, be no more than the shape of current containing dirty pages rate threshold value in current containing dirty pages rate
Under state, the internal memory migration control module, which performs, continues iteration copy.
Preferably, it is described interior in the state of continuing iteration copy number and reaching threshold value in the continuation iteration copy
Deposit migration control module and perform shutdown copy.
Preferably, in the continuation iteration copy, in the state of current containing dirty pages rate is less than current containing dirty pages rate threshold value, institute
State internal memory migration control module and perform shutdown copy.
Preferably, the internal memory migration control module preferentially copies the source void of low containing dirty pages rate in the continuation iteration copy
Plan machine internal memory, remaining source virtual machine internal memory is disposably copied in copy is shut down.
Compared with prior art, the present invention has following beneficial effect：
When the present invention can effectively reduce bulk migration relative to traditional pre-copy and rear copy internal memory migration mechanism
Between, total internal storage data migration amount greatly reduces；Meanwhile the present invention is realized for the adaptive void under different application scene
Plan machine scheduling of resource.
Brief description of the drawings
The detailed description made by reading with reference to the following drawings to non-limiting example, further feature of the invention,
Objects and advantages will become more apparent upon：
Fig. 1 is traditional virtual machine pre-copy and rear copy thermophoresis schematic diagram；
Fig. 2 is that the frame diagram of system is dispatched in the adaptive thermophoresis of virutal machine memory of the present invention
Fig. 3 is the adaptive thermophoresis scheduling flow figure of virutal machine memory of the present invention.
Embodiment
With reference to specific embodiment, the present invention is described in detail.Following examples will be helpful to the technology of this area
Personnel further understand the present invention, but the invention is not limited in any way.It should be pointed out that the ordinary skill to this area
For personnel, without departing from the inventive concept of the premise, some changes and improvements can also be made.These belong to the present invention
Protection domain.
Resources of virtual machine monitoring and scheduling system in a mixed cloud are built first, to realize to including internal memory, CPU, magnetic
The real-time monitoring of resources of virtual machine including disk and network bandwidth etc., system framework figure are as shown in Figure 2.Including to virtual
The monitoring of resource and scheduler module in machine.It is divided into Data Collection, data extraction on the whole and performs three root phases of scheduling.It is mixed
Closing in cloud has some physical machine cluster PM1,PM2,…,PMn, some virtual machines, such as PM are deployed in every physical machine1
On have virtual machine VM11,VM12,…,VM1i, and by monitor of virtual machine VMM configurations, manage and monitor these virtual machines.Data are received
The related data of virtual machine is monitored and collected with storage system modules by Data Collection by the collection stage；Hereafter virtual machine is adjusted
Degree platform extracts the virtual machine information of storage, by including virutal machine memory containing dirty pages rate RdPredicted Deng data analysis,
The process such as PM load estimations and internal memory migration control, form for the adaptive internal memory heat of virtual machine under current application situation
Migration strategy, finally using the virtual machine adaptive scheduling in this strategy execution mixed cloud, to reach the optimization of migration performance,
Realize load balancing simultaneously：When some physical machines are overweight or break down, reliably operating virtual machine can be moved
Move on in the physical machine of other normal uses, avoid the interruption of business in virtual machine；Or when some physical machine underloads, can
So that virtual machine to be integrated by migration, so as to reduce the quantity of physical machine, while saving and reducing expense, effectively
Improve the utilization rate of resource.
Time series autoregression model (hereinafter referred to as AR models) is with the process for itself doing regression variable, that is, utilizes early stage
Come the linear regression model (LRM) of certain moment stochastic variable after describing, it is time sequence for the linear combination of the stochastic variable at some moment
A kind of common form in row.The present invention is by among the prediction of AR model uses to virutal machine memory containing dirty pages rate, by applying
The comparison and judgement of internal memory containing dirty pages, the most suitable internal memory thermophoresis strategy of adaptive selection.Specific solution is as follows：
1st, internal memory containing dirty pages rate forecast model is built first.Predict virutal machine memory using AR Self-regression Forecast Models
Containing dirty pages rate R after t iteration copyd,tSituation, by this predicted value, the internal memory in virtual machine in mixed cloud PM can be made
Detected in real time with situation, Rd,tIt can be defined as：
Wherein, c is a constant, εtIt is the stochastic variable (white noise) that an average is zero,It is nearest
K (1≤k≤t-1) iteration copy model parameter, Rd,t-1,Rd,t-2,…,Rd,t-kIt is its corresponding internal memory containing dirty pages rate.According to
This understands Rd,tIt is k rank AR (k) autoregression model.
AR (k) autoregression models can also be represented with lag operator (Lag operator) L.If operator operation be by
The value of one time series previous moment is converted into currency, then this operator is referred to as lag operator, is designated as L.L can be defined
For：
LiRd,t=Rd,t-i
Wherein, LiL is designated as the i-th rank lag operator, such as three rank lag operators3。
So AR (k) autoregression models can be described as：
Wherein,It is AR lag operator multinomials, its expression formula is：
Therefore, containing dirty pages rate Rd,tExpression formula can be described as：
Wherein, φ (L) is a k rank lag operator multinomial, φ1,φ2,…,φkIt is model parameter.μ is one and is based on
Constant c and model parameter constant parameter variant.So far, virutal machine memory forecast model has been built up completing.
2nd, next it is the realization of internal memory adaptive scheduler strategy.It is adaptively hot that Fig. 3 illustrates whole virutal machine memory
The basic procedure of migration scheduling.It is first begin to the iteration copy stage.System will perform an iteration copy function, will be most virtual
Machine memory pages move to purpose virtual machine.Now, internal memory containing dirty pages produce in source virtual machine.Its quantity can be according in virtual machine
The application type of operation it is different and different.Current containing dirty pages rate RdCan be by above-mentioned AR (k) autoregression models come approximate
Predict and.
3rd, next by the containing dirty pages rate R of predictiondCompared with current containing dirty pages rate threshold value d.This threshold value is a dynamic
Changing value, its can according to applied in current virtual machine or business difference and change.Assuming that operation has m kinds in current source virtual machine
Using their current containing dirty pages rates areTherefore we define current containing dirty pages rate threshold value d and are
It is recognised that containing dirty pages rate threshold value d is the average value of the containing dirty pages rate of a different application, can approximately represent accordingly
Go out the state of internal memory in source virtual machine under current time.
4th, afterwards according to RdWith d comparative results, system by adaptive selection decide whether continue iteration copy or it is straight
Tap into and copied into shutdown and copy the stage on demand, two kinds of situations can be described as.If now Rd>D, i.e. internal memory containing dirty pages rate exceed
Present threshold value, then mean that source virtual machine has higher memory usage and frequently the page is read and write.So system will perform
Copy into shutdown copy and on demand the stage：Source virtual machine is hung up, starts purpose virtual machine, then on demand from source virtual machine
Memory pages are copied, this process copies migration after being similar to；
If the 5th, now Rd≤ d, that is, mean that internal memory containing dirty pages rate is not above present threshold value, system will continue executing with iteration
Copy.At this stage, we can set the containing dirty pages rate threshold value d ' after iteration each time：
The present invention introduces containing dirty pages rate comparison mechanism for each iteration copy, goes to identify the high containing dirty pages rate run in virtual machine
Using or service.If for example,Show that j-th of application is high containing dirty pages rate application, its memory pages has very maximum probability
It can be read and write in ensuing iteration, then high probability can will be retransmitted repeatedly in ensuing copy accordingly.Institute
So that in order to avoid retransmitting the internal memory of the application of this type repeatedly, containing dirty pages rate comparison mechanism can set priority, and prioritised transmission is low dirty
The internal memory of page rate application, when current containing dirty pages rate is less than current containing dirty pages rate threshold value or iterations and reaches threshold value n, into stopping
Machine copies the stage, disposably copies high containing dirty pages rate application memory and remaining status information to purpose virtual machine in the lump.Its
In, iterations threshold value n is rule of thumb and the empirical value obtained by the historical record of passing migration, general n take 10 times.Extremely
This, that is, complete the adaptive thermophoresis scheduling strategy of virutal machine memory.
The present invention pushes away to optimize live migration of virtual machine time and migrating data amount to three kinds of basic steps of memory copying
Send copy, shut down copy and on demand copy refined, and introduce internal memory containing dirty pages rate and iteration copy number measure, it is right
Internal memory on source virtual machine first carries out an iteration copy procedure, then the comparison by current containing dirty pages rate and containing dirty pages rate threshold value, certainly
The fixed iteration copy that whether continues still is directly entered shutdown copy and copies the stage on demand：(1) if continuing the iteration copy stage,
Then according to internal memory iterations or containing dirty pages rate compared with the containing dirty pages rate threshold value of current predictive, it is determined whether enter and shut down copy rank
Section；(2) if not continuing the iteration copy stage, it is directly entered and shuts down the copy stage, it is empty to purpose copies complete memory mirror
Plan machine, the stage is then being copied on demand on demand from source virtual machine copy internal memory, so as to be finally completed whole internal memory from source virtual machine
To the dynamic migration of purpose virtual machine.Algorithm is realized according to application different in virtual machine, different for internal memory service condition
And the process of the adaptive most suitable migration strategy of selection.Meanwhile the present invention utilizes time series Self-regression Forecast Model (AR moulds
Type) virutal machine memory containing dirty pages rate is predicted, it is next to predict by the containing dirty pages rate of memory copying iteration several times before
The virutal machine memory containing dirty pages situation at individual moment, and Dynamic comparison mechanism is introduced accordingly, by different application in current time virtual machine
Internal memory containing dirty pages rate compared with setting containing dirty pages rate threshold value, mark the memory pages higher than threshold value, only in last time
These pages are copied to purpose virtual machine when depositing iteration copy, are retransmitted repeatedly so as to avoid internal memory, optimize virtual machine heat
The transmission time and migrating data amount of migration.
One skilled in the art will appreciate that except realizing system provided by the invention in a manner of pure computer readable program code
And its beyond each device, module, unit, completely can be by the way that method and step progress programming in logic be provided come the present invention
System and its each device, module, unit with gate, switch, application specific integrated circuit, programmable logic controller (PLC) and embedding
Enter the form of the controller that declines etc. to realize identical function.So system provided by the invention and its every device, module, list
Member is considered a kind of hardware component, and is used to realize that device, module, the unit of various functions also may be used to what is included in it
To be considered as the structure in hardware component；It both can be real that will can also be considered as device, module, the unit of realizing various functions
The software module of existing method can be the structure in hardware component again.
The specific embodiment of the present invention is described above.It is to be appreciated that the invention is not limited in above-mentioned
Particular implementation, those skilled in the art can make a variety of changes or change within the scope of the claims, this not shadow
Ring the substantive content of the present invention.In the case where not conflicting, the feature in embodiments herein and embodiment can any phase
Mutually combination.
Claims (10)
- A kind of 1. adaptive thermophoresis dispatching method of virutal machine memory mixed under cloud mode, it is characterised in that including：Containing dirty pages rate forecast model construction step：Structure containing dirty pages rate forecast model carrys out prediction source virutal machine memory after iteration copy Current containing dirty pages rate；An iteration copies step：An iteration copy is carried out to source virtual machine internal memory, passes through current containing dirty pages rate and current containing dirty pages The comparison of rate threshold value, determine that performing continuation iteration copy step still performs shutdown copy and copy on demand；Continue iteration copy step：Source virtual machine internal memory is carried out to continue iteration copy, according to the number for continuing iteration copy, or Person decides whether to enter and shuts down copy according to the comparison of current containing dirty pages rate and current containing dirty pages rate threshold value；Wherein, current containing dirty pages rate threshold value is equal to the average value of the current containing dirty pages rate of all applications in source virtual machine internal memory.
- 2. the adaptive thermophoresis dispatching method of virutal machine memory under mixing cloud mode according to claim 1, its feature It is, in an iteration copies step, in the state of current containing dirty pages rate exceedes current containing dirty pages rate threshold value, performs shutdown Copy and copy on demand, in the state of current containing dirty pages rate is no more than current containing dirty pages rate threshold value, performs continuation iteration and copy step.
- 3. the adaptive thermophoresis dispatching method of virutal machine memory under mixing cloud mode according to claim 1, its feature It is, in the continuation iteration copy step, in the state of continuing iteration copy number and reaching threshold value, performs shutdown and copy Shellfish.
- 4. the adaptive thermophoresis dispatching method of virutal machine memory under mixing cloud mode according to claim 1, its feature It is, in the continuation iteration copy step, in the state of current containing dirty pages rate is less than current containing dirty pages rate threshold value, performs shutdown Copy.
- 5. the adaptive thermophoresis dispatching method of virutal machine memory under mixing cloud mode according to claim 1, its feature It is, the source virtual machine internal memory of low containing dirty pages rate is preferentially copied in the continuation iteration copy step, in copy is shut down once Property the remaining source virtual machine internal memory of copy.
- A kind of 6. adaptive thermophoresis scheduling system of virutal machine memory mixed under cloud mode, it is characterised in that including：Containing dirty pages rate forecast model：Current containing dirty pages rate of the prediction source virutal machine memory after iteration copy；Internal memory migration control module：An iteration copy is carried out to source virtual machine internal memory, passes through current containing dirty pages rate and current containing dirty pages The comparison of rate threshold value, determine to perform and continue iteration copy or perform to shut down copy and copy on demand；Copied performing continuation iteration During shellfish, according to the number for continuing iteration copy, or the comparison according to current containing dirty pages rate and current containing dirty pages rate threshold value, determine Whether enter and shut down copy；Wherein, current containing dirty pages rate threshold value is equal to the average value of the current containing dirty pages rate of all applications in source virtual machine internal memory.
- 7. the adaptive thermophoresis scheduling system of virutal machine memory under mixing cloud mode according to claim 6, its feature It is, in an iteration copy, in the state of current containing dirty pages rate exceedes current containing dirty pages rate threshold value, the internal memory migration Control module, which performs, shuts down copy and copies on demand, described in the state of current containing dirty pages rate is no more than current containing dirty pages rate threshold value Internal memory migration control module, which performs, continues iteration copy.
- 8. the adaptive thermophoresis scheduling system of virutal machine memory under mixing cloud mode according to claim 6, its feature It is, in the continuation iteration copy, in the state of continuing iteration copy number and reaching threshold value, the internal memory migration control Module, which performs, shuts down copy.
- 9. the adaptive thermophoresis scheduling system of virutal machine memory under mixing cloud mode according to claim 6, its feature It is, in the continuation iteration copy, in the state of current containing dirty pages rate is less than current containing dirty pages rate threshold value, the internal memory migration Control module, which performs, shuts down copy.
- 10. the adaptive thermophoresis scheduling system of virutal machine memory under mixing cloud mode according to claim 6, its feature It is, the internal memory migration control module preferentially copies the source virtual machine internal memory of low containing dirty pages rate in the continuation iteration copy, Remaining source virtual machine internal memory is disposably copied in copy is shut down.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710597298.2A CN107479944B (en) | 2017-07-20 | 2017-07-20 | Virtual machine memory self-adaptive thermal migration scheduling method and system in hybrid cloud mode |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710597298.2A CN107479944B (en) | 2017-07-20 | 2017-07-20 | Virtual machine memory self-adaptive thermal migration scheduling method and system in hybrid cloud mode |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107479944A true CN107479944A (en) | 2017-12-15 |
| CN107479944B CN107479944B (en) | 2021-11-30 |
Family
ID=60596762
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710597298.2A Active CN107479944B (en) | 2017-07-20 | 2017-07-20 | Virtual machine memory self-adaptive thermal migration scheduling method and system in hybrid cloud mode |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107479944B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110134490A (en) * | 2018-02-08 | 2019-08-16 | 中兴通讯股份有限公司 | Dynamic migration of virtual machine method, equipment and storage medium |
| CN110737509A (en) * | 2019-10-18 | 2020-01-31 | 杭州朗和科技有限公司 | Thermal migration processing method and device, storage medium and electronic equipment |
| CN110928636A (en) * | 2018-09-19 | 2020-03-27 | 阿里巴巴集团控股有限公司 | Virtual machine live migration method, device and equipment |
| CN111176781A (en) * | 2019-08-23 | 2020-05-19 | 腾讯科技（深圳）有限公司 | Virtual machine live migration method and device, storage medium and electronic equipment |
| WO2021212784A1 (en) * | 2020-04-24 | 2021-10-28 | 杭州网银互联科技股份有限公司 | Distributed storage-based live migration system and migration method therefor |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102185774A (en) * | 2011-05-10 | 2011-09-14 | 中兴通讯股份有限公司 | Method, monitor and system for seamless migration of virtual machine |
| CN102929715A (en) * | 2012-10-31 | 2013-02-13 | 曙光云计算技术有限公司 | Method and system for scheduling network resources based on virtual machine migration |
| CN103049308A (en) * | 2012-11-30 | 2013-04-17 | 华为技术有限公司 | Method, device and system for achieving thermal migration of virtual machine |
| CN103246561A (en) * | 2012-11-16 | 2013-08-14 | 佳都新太科技股份有限公司 | Real-time virtual machine shifting technology based on XEN |
| CN103955399A (en) * | 2014-04-30 | 2014-07-30 | 华为技术有限公司 | Migrating method and device for virtual machine, as well as physical host |
| US8924965B2 (en) * | 2011-01-07 | 2014-12-30 | Red Hat Israel, Ltd. | Memory state transfer of virtual machine-controlled peripherals during migrations of the virtual machine |
| CN104268003A (en) * | 2014-09-30 | 2015-01-07 | 南京理工大学 | Memory state migration method applicable to dynamic migration of virtual machine |
| KR20160099365A (en) * | 2015-02-12 | 2016-08-22 | 한국전자통신연구원 | Apparatus and method for memory migration based on omitting redundant in virtial machine migration |
| CN106775949A (en) * | 2016-12-28 | 2017-05-31 | 广西大学 | A kind of Application of composite feature that perceives migrates optimization method online with the virtual machine of the network bandwidth |
-
2017
- 2017-07-20 CN CN201710597298.2A patent/CN107479944B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8924965B2 (en) * | 2011-01-07 | 2014-12-30 | Red Hat Israel, Ltd. | Memory state transfer of virtual machine-controlled peripherals during migrations of the virtual machine |
| CN102185774A (en) * | 2011-05-10 | 2011-09-14 | 中兴通讯股份有限公司 | Method, monitor and system for seamless migration of virtual machine |
| CN102929715A (en) * | 2012-10-31 | 2013-02-13 | 曙光云计算技术有限公司 | Method and system for scheduling network resources based on virtual machine migration |
| CN103246561A (en) * | 2012-11-16 | 2013-08-14 | 佳都新太科技股份有限公司 | Real-time virtual machine shifting technology based on XEN |
| CN103049308A (en) * | 2012-11-30 | 2013-04-17 | 华为技术有限公司 | Method, device and system for achieving thermal migration of virtual machine |
| EP2916217A1 (en) * | 2012-11-30 | 2015-09-09 | Huawei Technologies Co., Ltd. | Method, apparatus, and system for implementing hot migration of virtual machine |
| CN103955399A (en) * | 2014-04-30 | 2014-07-30 | 华为技术有限公司 | Migrating method and device for virtual machine, as well as physical host |
| CN104268003A (en) * | 2014-09-30 | 2015-01-07 | 南京理工大学 | Memory state migration method applicable to dynamic migration of virtual machine |
| KR20160099365A (en) * | 2015-02-12 | 2016-08-22 | 한국전자통신연구원 | Apparatus and method for memory migration based on omitting redundant in virtial machine migration |
| CN106775949A (en) * | 2016-12-28 | 2017-05-31 | 广西大学 | A kind of Application of composite feature that perceives migrates optimization method online with the virtual machine of the network bandwidth |
Non-Patent Citations (3)
| Title |
|---|
| HELLEN MAZIKU等: "Network Aware VM Migration in Cloud Data Centers", 《2014 THIRD GENI RESEARCH AND EDUCATIONAL EXPERIMENT WORKSHOP》 * |
| 常德成等: "虚拟机动态迁移方法", 《计算机应用研究》 * |
| 邹潇: "基于脏页率预测的虚拟机动态迁移研究", 《中国优秀硕士论文全文库 信息科技辑》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110134490A (en) * | 2018-02-08 | 2019-08-16 | 中兴通讯股份有限公司 | Dynamic migration of virtual machine method, equipment and storage medium |
| CN110134490B (en) * | 2018-02-08 | 2023-12-29 | 中兴通讯股份有限公司 | Virtual machine dynamic migration method, device and storage medium |
| CN110928636A (en) * | 2018-09-19 | 2020-03-27 | 阿里巴巴集团控股有限公司 | Virtual machine live migration method, device and equipment |
| CN111176781A (en) * | 2019-08-23 | 2020-05-19 | 腾讯科技（深圳）有限公司 | Virtual machine live migration method and device, storage medium and electronic equipment |
| CN110737509A (en) * | 2019-10-18 | 2020-01-31 | 杭州朗和科技有限公司 | Thermal migration processing method and device, storage medium and electronic equipment |
| CN110737509B (en) * | 2019-10-18 | 2022-04-19 | 杭州朗和科技有限公司 | Thermal migration processing method and device, storage medium and electronic equipment |
| WO2021212784A1 (en) * | 2020-04-24 | 2021-10-28 | 杭州网银互联科技股份有限公司 | Distributed storage-based live migration system and migration method therefor |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107479944B (en) | 2021-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107479944A (en) | Mix the adaptive thermophoresis dispatching method of virutal machine memory and system under cloud mode | |
| CN103955399B (en) | A kind of virtual machine migration method and device and physical host | |
| CN109491790B (en) | Container-based industrial Internet of things edge computing resource allocation method and system | |
| JP7222029B2 (en) | Holistic global performance and power management | |
| CN103577249B (en) | The online moving method of virtual machine and system | |
| CN102156665B (en) | Differential serving method for virtual system competition resources | |
| CN103365704B (en) | Internal memory pre-copy method in virtual machine (vm) migration and perform device and the system of the method | |
| CN105302630B (en) | A kind of dynamic adjusting method and its system of virtual machine | |
| CN108038040A (en) | Computer cluster performance indicator detection method, electronic equipment and storage medium | |
| WO2013082119A1 (en) | Cloud provisioning accelerator | |
| CN104991830A (en) | YARN resource allocation and energy-saving scheduling method and system based on service level agreement | |
| CN102184125A (en) | Load balancing method based on program behaviour online analysis under heterogeneous multi-core environment | |
| CN113806018B (en) | Kubernetes cluster resource mixed scheduling method based on neural network and distributed cache | |
| CN102739785B (en) | Method for scheduling cloud computing tasks based on network bandwidth estimation | |
| CN109408220A (en) | A kind of task processing method and device | |
| CN107404409A (en) | Towards the container cloud elastic supply number of containers Forecasting Methodology and system of mutation load | |
| CN104935628A (en) | Method of migrating a plurality of associated virtual machines among a plurality of data centers | |
| CN106230986A (en) | The resource adaptation dispatching patcher of a kind of electrically-based PaaS cloud platform and method | |
| CN106775949A (en) | A kind of Application of composite feature that perceives migrates optimization method online with the virtual machine of the network bandwidth | |
| Wang et al. | Predicting CPU usage for proactive autoscaling | |
| CN106201700A (en) | The dispatching method that a kind of virtual machine migrates online | |
| Tran et al. | Proactive stateful fault-tolerant system for kubernetes containerized services | |
| Li et al. | Load prediction-based automatic scaling cloud computing | |
| Venkataswamy et al. | Rare: Renewable energy aware resource management in datacenters | |
| EP4189542A1 (en) | Sharing of compute resources between the virtualized radio access network (vran) and other workloads |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |