CN110119301A - A kind of virtual machine migration method and system - Google Patents
A kind of virtual machine migration method and system Download PDFInfo
- Publication number
- CN110119301A CN110119301A CN201810124279.2A CN201810124279A CN110119301A CN 110119301 A CN110119301 A CN 110119301A CN 201810124279 A CN201810124279 A CN 201810124279A CN 110119301 A CN110119301 A CN 110119301A
- Authority
- CN
- China
- Prior art keywords
- virtual machine
- resource
- host
- hot spot
- metrics
- 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
- 230000005012 migration Effects 0.000 title claims abstract description 54
- 238000013508 migration Methods 0.000 title claims abstract description 54
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000001816 cooling Methods 0.000 claims description 16
- 239000011800 void material Substances 0.000 claims description 4
- 238000012545 processing Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 230000002567 autonomic effect Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
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
- 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
- G06F9/44—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
- G06F2009/4557—Distribution of virtual machine instances; Migration and load balancing
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention discloses a kind of virtual machine migration methods, comprising: identifies hot spot host according to the resource service condition of each host；Wherein, the resource service condition includes four resource metrics values, respectively network I/O, CPU usage, memory usage and disk I/O；Identify the corresponding hot spot virtual machine of each resource metrics in the hot spot host；The corresponding purpose host of each resource metrics is chosen from preset resource pool；It will be on the corresponding hot spot virtual machine (vm) migration to corresponding purpose host of resource metrics each in the hot spot host.The invention also discloses a kind of virtual machine (vm) migration systems.The present invention can accurately in real time migrate virtual machine, improve resource utilization.
Description
Technical field
The present invention relates to field of computer technology more particularly to a kind of virtual machine migration method and systems.
Background technique
Currently, under virtualized environment, the loading condition of virtual machine can not be generally predicted in advance, if in the process of running,
Multiple virtual machine loads are all very high on certain hosts, the case where resource is fought for occur, and the load of other host is lower,
So in order to not influence business processing, need partial virtual machine moving to the lower host of load from higher host is loaded
On machine.
The processing method of the prior art is that O&M engineer first has found such case, then goes to look for load higher and lower
Then host executes migration by hand.Simultaneously in enterprise's private clound, if be on a grand scale, there are many operation system, can inevitably go out
Existing business side is not destroyed in time when not using virtual machine, is caused the presence of the virtual machine not used much in cluster, is accounted for
With a large amount of resource.
Summary of the invention
The present invention aiming at the problems existing in the prior art, provides a kind of virtual machine migration method and system, Neng Gouzhun
Really virtual machine is migrated in real time, improves resource utilization.
The technical solution that the present invention is proposed with regard to above-mentioned technical problem is as follows:
On the one hand, the present invention provides a kind of virtual machine migration method, comprising:
Hot spot host is identified according to the resource service condition of each host；Wherein, the resource service condition includes
Four resource metrics values, respectively network I/O, CPU usage, memory usage and disk I/O；
Identify the corresponding hot spot virtual machine of each resource metrics in the hot spot host；
The corresponding purpose host of each resource metrics is chosen from preset resource pool；
By the corresponding hot spot virtual machine (vm) migration of resource metrics each in the hot spot host to corresponding purpose host
On.
Further, described that hot spot host is identified according to the resource service condition of each host, it specifically includes:
The resource service condition of each host of circle collection；
Data of every acquisition, then by the corresponding pre-set level threshold value of each resource metrics value of the host into
Row compares；
If same class resource metrics value reaches its corresponding metrics-thresholds in the data of continuous n times acquisition, will be described
Host is identified as hot spot host；Wherein, N >=1.
Further, the corresponding hot spot virtual machine of each resource metrics in the identification hot spot host, it is specific to wrap
It includes:
Acquire the resource service condition of each virtual machine on the hot spot host；
Respectively using each maximum virtual machine of resource metrics value as hot spot virtual machine.
Further, described before choosing the corresponding purpose host of each resource metrics in preset resource pool,
Further include:
The resource pool of each resource metrics is set；
Each resource pool is divided at least two child resource ponds, and different child resources pond corresponds to the resource of different range and refers to
Scale value；
According to the resource metrics value of each host, the host is added separately to the child resource of corresponding resource index
Chi Zhong.
Further, described that the corresponding purpose host of each resource metrics is chosen from preset resource pool, it is specific to wrap
It includes:
Hot spot virtual machine corresponding for each resource metrics is chosen according to the child resource pond of resource metrics value from low to high
Sequentially, the host different from the hot spot virtual machine business is chosen from the child resource pond, and will be in the virtual machine of selection
The minimum virtual machine of resource metrics value host as a purpose.
Further, the virtual machine migration method further include:
Low-load virtual machine is identified according to the resource service condition of each virtual machine；
The low-load virtual machine is destroyed.
Further, described that low-load virtual machine is identified according to the resource service condition of each virtual machine, it specifically includes:
The virtual machine of low-load state is according to the identification of the resource service condition of each virtual machine；
The virtual machine being not used by the virtual machine in low-load state is identified as low-load virtual machine.
Further, the virtual machine that low-load state is according to the identification of the resource service condition of each virtual machine,
It specifically includes:
The resource service condition of each virtual machine of circle collection；
Data of every acquisition, then by the corresponding pre-set level threshold value of each resource metrics value of the virtual machine into
Row compares；
It, will be described if same class resource metrics value is below its corresponding metrics-thresholds in the data of continuous N time acquisition
Virtual machine is identified as the virtual machine in low-load state；Wherein, M >=1.
Further, described that the virtual machine being not used by the virtual machine in low-load state is identified as low bear
Virtual machine is carried, is specifically included:
Virtual machine in low-load state is put into cooling bay；
Prompting message is sent to user terminal by the virtual machine in the cooling bay；
If receiving the information of the client feeds back in preset duration, determine that the virtual machine is still used, it will
The virtual machine is removed from cooling bay；
If not receiving the information of the client feeds back in preset duration, determine that the virtual machine is not used by,
The virtual machine is identified as low-load virtual machine.
On the other hand, the present invention provides a kind of virtual machine (vm) migration system, can be realized the institute of above-mentioned virtual machine migration method
There is process, the virtual machine (vm) migration system includes:
Hot spot host identification module, for identifying hot spot host according to the resource service condition of each host；Its
In, the resource service condition includes four resource metrics values, respectively network I/O, CPU usage, memory usage and disk
IO；
Hot spot virtual machine identification module, the corresponding hot spot of each resource metrics is virtual in the hot spot host for identification
Machine；
Purpose host chooses module, for choosing the corresponding purpose host of each resource metrics from preset resource pool
Machine；And
Hot spot virtual machine (vm) migration module is used for the corresponding hot spot virtual machine of resource metrics each in the hot spot host
It moves on corresponding purpose host.
Technical solution provided in an embodiment of the present invention has the benefit that
Obtained by this four resource metrics values of network I/O, CPU usage, memory usage and disk I/O host and
The resource service condition of virtual machine, to identify the corresponding hot spot virtual machine of each resource metrics in hot spot host, in turn
Chosen from resource pool and load lower host host as a purpose, so as to by hot spot virtual machine Autonomic Migration Framework to accordingly
On purpose host, without artificial migration, O&M difficulty is greatly reduced, saving people can cost；Low-load virtual machine is carried out
Identification and destruction, effectively improve resource utilization, avoid the unnecessary wasting of resources.
Detailed description of the invention
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for
For those of ordinary skill in the art, without creative efforts, it can also be obtained according to these attached drawings other
Attached drawing.
Fig. 1 is the flow diagram for the virtual machine migration method that the embodiment of the present invention one provides；
Fig. 2 is the structural schematic diagram of virtual machine (vm) migration system provided by Embodiment 2 of the present invention；
Fig. 3 is the concrete structure schematic diagram for the virtual machine (vm) migration system that the embodiment of the present invention three provides.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to embodiment party of the present invention
Formula is described in further detail.
Embodiment one
The embodiment of the invention provides a kind of virtual machine migration methods, referring to Fig. 1, this method comprises:
S1, hot spot host is identified according to the resource service condition of each host；Wherein, the resource service condition packet
Include four resource metrics values, respectively network I/O, CPU usage, memory usage and disk I/O；
The corresponding hot spot virtual machine of each resource metrics in S2, the identification hot spot host；
S3, the corresponding purpose host of each resource metrics is chosen from preset resource pool；
S4, by the corresponding hot spot virtual machine (vm) migration of resource metrics each in the hot spot host to corresponding purpose host
On machine.
It should be noted that in virtual machine environment, although virtual machine traffic pressure can be estimated by investigating in advance,
Certain scheduling strategy is taken, so that the big virtual machine of traffic pressure spreads out as far as possible, avoids the occurrence of the case where resource is fought for, but
Most of business scenario is changeable in reality, pressure is different, can not predict traffic pressure in advance, so estimating pressure in advance and taking
The scheme of scheduling strategy can only solve subproblem.And the present embodiment can make according to collected network I/O, disk I/O, CPU
With real time datas such as rate, memory usages, comprehensively considered according to preset recognition strategy from multiple dimensions, is sent out comprehensively in time
Hot point resource in existing cluster, and early warning is generated, while the precondition as next step automation migration.The present embodiment is directed to
It was found that hot point resource automation move to the lower host of load, realize full-automatic O&M, greatly reduce O&M hardly possible
Degree, saving people can cost.
Specifically, in step sl, described that hot spot host is identified according to the resource service condition of each host, specifically
Include:
The resource service condition of each host of circle collection；
Data of every acquisition, then by the corresponding pre-set level threshold value of each resource metrics value of the host into
Row compares；
If same class resource metrics value reaches its corresponding metrics-thresholds in the data of continuous n times acquisition, will be described
Host is identified as hot spot host；Wherein, N >=1.
It should be noted that user can be network I/O, CPU usage, four memory usage, disk I/O resource metrics
Threshold value, i.e. metrics-thresholds are set, while the identification condition of hot spot host can also be set, which can be single resource and refer to
Scale value reaches its corresponding metrics-thresholds, is also possible to any number of resource metrics values and reaches its corresponding metrics-thresholds.In addition,
The collection period of the resource service condition of also settable host, if (value of N can be according to system by the continuous N of some host
Tolerance is arranged, and the bigger expression of N can tolerate the high load of host longer time) a period data collected reach and set
Fixed hot spot host identifies condition, then the host is identified as hot spot host.
Specifically, in step s 2, the corresponding hot spot of each resource metrics is virtual in the identification hot spot host
Machine specifically includes:
Acquire the resource service condition of each virtual machine on the hot spot host；
Respectively using each maximum virtual machine of resource metrics value as hot spot virtual machine.
It should be noted that need to only identify hot spot virtual machine on hot spot host.For each hot spot host, traversal
All virtual machines thereon find out index from four network I/O, CPU usage, memory usage, disk I/O resource metrics respectively
It is worth the maximum virtual machine of ranking to make, forms the list of hot spot virtual machine.
Further, before step S3, i.e., it is corresponding that each resource metrics are chosen from preset resource pool described
Before purpose host, further includes:
The resource pool of each resource metrics is set；
Each resource pool is divided at least two child resource ponds, and different child resources pond corresponds to the resource of different range and refers to
Scale value；
According to the resource metrics value of each host, the host is added separately to the child resource of corresponding resource index
Chi Zhong.
It should be noted that for four network I/O, CPU usage, memory usage, disk I/O resource metrics settings four
The resource pool of seed type is respectively as follows: cpu resource pond, Internet resources pond, memory source pond, disk resource pond.For every kind of resource
The child resource pond of three levels of high, normal, basic utilization rate is generally arranged in pond, according to the resource service condition on host by host
One be added in high, normal, basic three levels, such as CPU usage are added to high utilization rate in 70% or more host
CPU child resource pond, CPU usage utilization rate CPU child resource pond, CPU usage in the host of 30%-70% is added to exist
30% host below is added to low utilization rate CPU child resource pond.The purpose for the arrangement is that in order to make the apparent appearance of administrator
The easy resource service condition for checking each host in system, while also weight is provided for the empty machine on subsequent migration hot spot host
The information wanted can very easily design scheduling strategy according to this information.
Specifically, in step s3, described that the corresponding purpose host of each resource metrics is chosen from preset resource pool
Machine specifically includes:
Hot spot virtual machine corresponding for each resource metrics is chosen according to the child resource pond of resource metrics value from low to high
Sequentially, the host different from the hot spot virtual machine business is chosen from the child resource pond, and will be in the virtual machine of selection
The minimum virtual machine of resource metrics value host as a purpose.
It should be noted that the resource metrics according to belonging to hot spot virtual machine, find the resource pool for the type that is corresponding to it, into
And according to low utilization rate, middle utilization rate, the sequence of high utilization rate, a host is selected from child resource pond as migration mesh
Host.Wherein, when choosing purpose host, the low host of respective resources index value is chosen, for example, hot spot host
On network I/O load highest, then just as far as possible that the highest hot spot scheduling virtual machine of network I/O load is low to network I/O load
On host, to achieve the purpose that host load balancing in cluster.In addition, purpose host also need meet belong to it is same
The virtual machine of a business cannot fall on this necessary condition on the same host, i.e., the industry of all virtual machines on purpose host
It is engaged in different from the business of hot spot virtual machine.
Further, the virtual machine migration method further include:
Low-load virtual machine is identified according to the resource service condition of each virtual machine；
The low-load virtual machine is destroyed.
It should be noted that propulsion of the cluster with the time, inevitably will appear some corpse resources not used.These
Corpse resource includes the network port, IP address, disk space, CPU core number, some resources are individually left, such as in pin
When ruining virtual machine, part resource is not discharged thoroughly, some are entire virtual machines, which does not bring tension management into
Platform perhaps virtual machine although bring management platform management into but unclaimed or user forget destroy or make
It is lower with rate, it needs to clear up these corpse resources at this time.The present embodiment is according to the resource service condition of virtual machine to void
Quasi- machine is identified, carries out automation thorough cleaning for the corpse resource that virtual machine is left, unclaimed, user is forgotten
The virtual machine that note is destroyed is destroyed, to improve resource utilization.
Specifically, described that low-load virtual machine is identified according to the resource service condition of each virtual machine, it specifically includes:
The virtual machine of low-load state is according to the identification of the resource service condition of each virtual machine；
The virtual machine being not used by the virtual machine in low-load state is identified as low-load virtual machine.
It should be noted that first being identified to the virtual machine in low-load state, for identifying in low negative
The virtual machine of load state also needs to judge its whether unclaimed or unmanned use, if unclaimed, determines that the virtual machine is low
Virtual machine is loaded, is directly destroyed.
Specifically, the virtual machine that low-load state is according to the identification of the resource service condition of each virtual machine, tool
Body includes:
The resource service condition of each virtual machine of circle collection；
Data of every acquisition, then by the corresponding pre-set level threshold value of each resource metrics value of the virtual machine into
Row compares；
It, will be described if same class resource metrics value is below its corresponding metrics-thresholds in the data of continuous N time acquisition
Virtual machine is identified as the virtual machine in low-load state；Wherein, M >=1.
It should be noted that the identification of the virtual machine in low-load state is similar with the identification of hot spot host.User
Threshold value can be set for four network I/O, CPU usage, memory usage, disk I/O resource metrics, i.e. metrics-thresholds are (low negative
Carry metrics-thresholds), while identification condition can also be set, which can be single resource metrics value lower than its corresponding index
Threshold value is also possible to any number of resource metrics values lower than its corresponding metrics-thresholds.In addition, the resource of also settable virtual machine
The collection period of service condition will if some virtual machine continuous N period data collected is lower than the identification condition of setting
The virtual machine is identified as the virtual machine in low-load state.
Specifically, described that the virtual machine being not used by the virtual machine in low-load state is identified as low-load
Virtual machine specifically includes:
Virtual machine in low-load state is put into cooling bay；
Prompting message is sent to user terminal by the virtual machine in the cooling bay；
If receiving the information of the client feeds back in preset duration, determine that the virtual machine is still used, it will
The virtual machine is removed from cooling bay；
If not receiving the information of the client feeds back in preset duration, determine that the virtual machine is not used by,
The virtual machine is identified as low-load virtual machine.
It should be noted that by virtual machine be added cooling bay when, to user terminal send the prompting messages such as mail and short message with
User is informed, if user has response to also continue moving the virtual machine from cooling bay using the virtual machine in the time at one end
Out, otherwise, continue to stay in cooling bay.Periodic scan is carried out to the virtual machine in cooling bay, virtual machine is often scanned once,
Count is incremented for it, and when its counting reaches setting value, which is destroyed.By cooling bay in low-load state
Virtual machine collected, i.e., can be quickly found out these void for being in low-load state without virtual machines whole in scanning system
Quasi- machine, reduces overhead.
The embodiment of the present invention by this four resource metrics values of network I/O, CPU usage, memory usage and disk I/O come
The resource service condition of host and virtual machine is obtained, to identify the corresponding hot spot of each resource metrics in hot spot host
Virtual machine, and then chosen from resource pool and load lower host host as a purpose, so that hot spot virtual machine is automatic
It moves on corresponding purpose host, without artificial migration, greatly reduces O&M difficulty, saving people can cost；To low negative
It carries virtual machine to be identified and destroyed, effectively improves resource utilization, avoid the unnecessary wasting of resources.
Embodiment two
The embodiment of the invention provides a kind of virtual machine (vm) migration systems, can be realized all of above-mentioned virtual machine migration method
Process, referring to fig. 2, the virtual machine (vm) migration system includes:
Hot spot host identification module 21, for identifying hot spot host according to the resource service condition of each host；
Wherein, the resource service condition includes four resource metrics values, respectively network I/O, CPU usage, memory usage and magnetic
Disk IO；
Hot spot virtual machine identification module 22, the corresponding hot spot of each resource metrics is empty in the hot spot host for identification
Quasi- machine；
Purpose host chooses module 23, for choosing the corresponding purpose place of each resource metrics from preset resource pool
Host；And
Hot spot virtual machine (vm) migration module 24, for the corresponding hot spot of resource metrics each in the hot spot host is virtual
Machine moves on corresponding purpose host.
Embodiment three
Referring to Fig. 3, it is the concrete structure schematic diagram of virtual machine (vm) migration system provided in an embodiment of the present invention, can be realized
State all processes of virtual machine migration method, the virtual machine (vm) migration system include achievement data acquisition module 31, database 32,
Resource identification module 33, resource processing module 34, migration process module 35 and destroy module 36.
There is virtual machine running in current cluster, each host is fitted with achievement data acquisition module 31.Refer to
It marks data acquisition module 31 and periodically acquires data from two dimensions of host and virtual machine, by collected host and void
The data such as network I/O, CPU usage, memory usage, the disk I/O of quasi- machine are stored into the database 32 of rear end.Meanwhile if
The recognition strategy of the recognition strategy, low-load virtual machine of setting hot spot host and hot spot virtual machine is also stored into the database of rear end
In 32.
Resource identification module 33 periodically obtains data from database 32, is identified according to hot spot host recognition strategy
Hot spot host out identifies that network I/O, CPU are used further according to hot spot virtual machine recognition strategy respectively from hot spot host
Hot spot virtual machine in this four index dimensions of rate, memory usage, disk I/O, and then by hot spot virtual machine information and its affiliated
Resource metrics information be sent to resource processing module 34.In addition, resource identification module 33 is identified also according to low-load virtual machine
Strategy identifies the virtual machine in low-load state, and the virtual machine in low-load state that will identify that is added to cooling
Pond, while mail and short message are sent to user terminal to inform user.
Resource processing module 34 is when receiving the information of the transmission of resource identification module 33, according to belonging to hot spot virtual machine
Resource metrics find the resource pool for the type that is corresponding to it, to select suitable purpose host.Picking out suitable purpose place
After host, hot spot host machine information and purpose host's machine information are sent to migration process module 35.In addition, resource processing module
Virtual machine in 34 pairs of cooling bays carries out periodic scan and is identified as when the scanning times of virtual machine reach predetermined value
Low-load virtual machine, and low-load virtual machine information is sent to and destroys module 36.
Hot spot host is moved to mesh after the information for receiving the transmission of resource processing module 34 by migration process module 35
Host on, for example, the virtual machine 1 in Fig. 3 in host 1 is moved on host 2, by the virtual machine 2 in host 2
It moves on host 3.
Module 36 is destroyed after the information for receiving the transmission of resource processing module 34, pin is carried out to low-load virtual machine
Processing is ruined, for example, destroying to the virtual machine 2 on the host 1 in Fig. 3.
The embodiment of the present invention by this four resource metrics values of network I/O, CPU usage, memory usage and disk I/O come
The resource service condition of host and virtual machine is obtained, to identify the corresponding hot spot of each resource metrics in hot spot host
Virtual machine, and then chosen from resource pool and load lower host host as a purpose, so that hot spot virtual machine is automatic
It moves on corresponding purpose host, without artificial migration, greatly reduces O&M difficulty, saving people can cost；To low negative
It carries virtual machine to be identified and destroyed, effectively improves resource utilization, avoid the unnecessary wasting of resources.
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all in spirit of the invention and
Within principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (10)
1. a kind of virtual machine migration method characterized by comprising
Hot spot host is identified according to the resource service condition of each host；Wherein, the resource service condition includes four
Resource metrics value, respectively network I/O, CPU usage, memory usage and disk I/O；
Identify the corresponding hot spot virtual machine of each resource metrics in the hot spot host；
The corresponding purpose host of each resource metrics is chosen from preset resource pool；
It will be on the corresponding hot spot virtual machine (vm) migration to corresponding purpose host of resource metrics each in the hot spot host.
2. virtual machine migration method as described in claim 1, which is characterized in that described to be used according to the resource of each host
Situation identifies hot spot host, specifically includes:
The resource service condition of each host of circle collection；
Data of every acquisition, then compare the corresponding pre-set level threshold value of each resource metrics value of the host
Compared with；
If same class resource metrics value reaches its corresponding metrics-thresholds in the data of continuous n times acquisition, by the host
Machine is identified as hot spot host；Wherein, N >=1.
3. virtual machine migration method as described in claim 1, which is characterized in that each in the identification hot spot host
The corresponding hot spot virtual machine of resource metrics, specifically includes:
Acquire the resource service condition of each virtual machine on the hot spot host；
Respectively using each maximum virtual machine of resource metrics value as hot spot virtual machine.
4. virtual machine migration method as described in claim 1, which is characterized in that chosen often from preset resource pool described
Before the corresponding purpose host of a resource metrics, further includes:
The resource pool of each resource metrics is set；
Each resource pool is divided at least two child resource ponds, and different child resources pond corresponds to the resource metrics of different range
Value；
According to the resource metrics value of each host, the host is added separately to the child resource pond of corresponding resource index
In.
5. virtual machine migration method as claimed in claim 4, which is characterized in that described to be chosen each from preset resource pool
The corresponding purpose host of resource metrics, specifically includes:
Hot spot virtual machine corresponding for each resource metrics is chosen suitable according to the child resource pond of resource metrics value from low to high
Sequence is chosen the host different from the hot spot virtual machine business from the child resource pond, and will be provided in the virtual machine of selection
The minimum virtual machine of source index value host as a purpose.
6. virtual machine migration method as described in claim 1, which is characterized in that the virtual machine migration method further include:
Low-load virtual machine is identified according to the resource service condition of each virtual machine；
The low-load virtual machine is destroyed.
7. virtual machine migration method as claimed in claim 6, which is characterized in that described to be used according to the resource of each virtual machine
Situation identifies low-load virtual machine, specifically includes:
The virtual machine of low-load state is according to the identification of the resource service condition of each virtual machine；
The virtual machine being not used by the virtual machine in low-load state is identified as low-load virtual machine.
8. virtual machine migration method as claimed in claim 7, which is characterized in that described to be used according to the resource of each virtual machine
Situation identification is in the virtual machine of low-load state, specifically includes:
The resource service condition of each virtual machine of circle collection；
Data of every acquisition, then compare the corresponding pre-set level threshold value of each resource metrics value of the virtual machine
Compared with；
It, will be described virtual if same class resource metrics value is below its corresponding metrics-thresholds in the data of continuous N time acquisition
Machine is identified as the virtual machine in low-load state；Wherein, M >=1.
9. virtual machine migration method as claimed in claim 7, which is characterized in that described by the void in low-load state
The virtual machine being not used by quasi- machine is identified as low-load virtual machine, specifically includes:
Virtual machine in low-load state is put into cooling bay；
Prompting message is sent to user terminal by the virtual machine in the cooling bay；
If receiving the information of the client feeds back in preset duration, determine that the virtual machine is still used, it will be described
Virtual machine is removed from cooling bay；
If not receiving the information of the client feeds back in preset duration, determine that the virtual machine is not used by, by institute
It states virtual machine and is identified as low-load virtual machine.
10. a kind of virtual machine (vm) migration system, can be realized virtual machine migration method as described in any one of claim 1 to 9,
It is characterized in that, the virtual machine (vm) migration system includes:
Hot spot host identification module, for identifying hot spot host according to the resource service condition of each host；Wherein, institute
Stating resource service condition includes four resource metrics values, respectively network I/O, CPU usage, memory usage and disk I/O；
Hot spot virtual machine identification module, for identification corresponding hot spot virtual machine of each resource metrics in the hot spot host；
Purpose host chooses module, for choosing the corresponding purpose host of each resource metrics from preset resource pool；
And
Hot spot virtual machine (vm) migration module is used for the corresponding hot spot virtual machine (vm) migration of resource metrics each in the hot spot host
Onto corresponding purpose host.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810124279.2A CN110119301A (en) | 2018-02-07 | 2018-02-07 | A kind of virtual machine migration method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810124279.2A CN110119301A (en) | 2018-02-07 | 2018-02-07 | A kind of virtual machine migration method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110119301A true CN110119301A (en) | 2019-08-13 |
Family
ID=67520138
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810124279.2A Pending CN110119301A (en) | 2018-02-07 | 2018-02-07 | A kind of virtual machine migration method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110119301A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113127146A (en) * | 2020-01-16 | 2021-07-16 | 上海盛霄云计算技术有限公司 | Heterogeneous dynamic random scheduling method and system |
| CN115878329A (en) * | 2023-02-02 | 2023-03-31 | 天翼云科技有限公司 | Host resource scheduling method and device, electronic equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102937912A (en) * | 2012-11-28 | 2013-02-20 | 华为技术有限公司 | Method and equipment for scheduling virtual machine |
| US20140082202A1 (en) * | 2012-08-21 | 2014-03-20 | Huawei Technologies Co., Ltd. | Method and Apparatus for Integration of Virtual Cluster and Virtual Cluster System |
| US20140107855A1 (en) * | 2011-12-29 | 2014-04-17 | Huawei Technologies Co., Ltd. | Energy Saving Monitoring Method and Device |
| CN105740074A (en) * | 2016-01-26 | 2016-07-06 | 中标软件有限公司 | Cloud computing based virtual machine load balancing method |
| CN106133693A (en) * | 2015-02-28 | 2016-11-16 | 华为技术有限公司 | The moving method of virtual machine, device and equipment |
| CN106227578A (en) * | 2016-07-12 | 2016-12-14 | 腾讯科技（深圳）有限公司 | A kind of method of live migration of virtual machine, equipment and system |
| CN106933650A (en) * | 2017-03-03 | 2017-07-07 | 北方工业大学 | load management method and system of cloud application system |
-
2018
- 2018-02-07 CN CN201810124279.2A patent/CN110119301A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140107855A1 (en) * | 2011-12-29 | 2014-04-17 | Huawei Technologies Co., Ltd. | Energy Saving Monitoring Method and Device |
| US20140082202A1 (en) * | 2012-08-21 | 2014-03-20 | Huawei Technologies Co., Ltd. | Method and Apparatus for Integration of Virtual Cluster and Virtual Cluster System |
| CN102937912A (en) * | 2012-11-28 | 2013-02-20 | 华为技术有限公司 | Method and equipment for scheduling virtual machine |
| CN106133693A (en) * | 2015-02-28 | 2016-11-16 | 华为技术有限公司 | The moving method of virtual machine, device and equipment |
| CN105740074A (en) * | 2016-01-26 | 2016-07-06 | 中标软件有限公司 | Cloud computing based virtual machine load balancing method |
| CN106227578A (en) * | 2016-07-12 | 2016-12-14 | 腾讯科技（深圳）有限公司 | A kind of method of live migration of virtual machine, equipment and system |
| CN106933650A (en) * | 2017-03-03 | 2017-07-07 | 北方工业大学 | load management method and system of cloud application system |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113127146A (en) * | 2020-01-16 | 2021-07-16 | 上海盛霄云计算技术有限公司 | Heterogeneous dynamic random scheduling method and system |
| CN115878329A (en) * | 2023-02-02 | 2023-03-31 | 天翼云科技有限公司 | Host resource scheduling method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11379272B2 (en) | Autoscaling using file access or cache usage for cluster machines | |
| EP2715539B1 (en) | Distributed computing framework | |
| US20150363238A1 (en) | Resource management in a virtualized computing environment | |
| CN105159768A (en) | Task management method and cloud data center management platform | |
| JP4636625B2 (en) | NIC connection control method for virtual network system, NIC connection control system for virtual network, and program | |
| CN104539708B (en) | A kind of capacity reduction method, device and the system of cloud platform resource | |
| CN104199739B (en) | A kind of speculating type Hadoop dispatching methods based on load balancing | |
| US20180267869A1 (en) | Method and apparatus for processing gateway device fault | |
| US10379906B2 (en) | Method and apparatus for system call command batch processing | |
| CN101595456A (en) | The method and system that is used for transaction resource control | |
| US9489210B2 (en) | Implementing configuration preserving relocation of SRIOV adapter | |
| CN107515781B (en) | Deterministic task scheduling and load balancing system based on multiple processors | |
| CN111190691A (en) | Automatic migration method, system, device and storage medium suitable for virtual machine | |
| CN109710414A (en) | A kind of job scheduling method, device, equipment and storage medium | |
| CN110727508A (en) | Task scheduling system and scheduling method | |
| CN114564302B (en) | GPU resource allocation method, system, equipment and medium | |
| CN110221936A (en) | Database alert processing method, device, equipment and computer readable storage medium | |
| CN115543577A (en) | Kubernetes resource scheduling optimization method based on covariates, storage medium and equipment | |
| CN110119301A (en) | A kind of virtual machine migration method and system | |
| CN115658311A (en) | Resource scheduling method, device, equipment and medium | |
| CN100373342C (en) | Method and system of thread synchronization in simultaneous multi-threaded processor machines | |
| CN110222284A (en) | Multi-page loading method and computer readable storage medium | |
| US20150286493A1 (en) | Virtual-machine placement based on information from multiple data centers | |
| CN107102966B (en) | Multi-core processor chip, interrupt control method and controller | |
| CN104735063A (en) | Security evaluating and detecting method used for cloud infrastructure |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190813 |
|
| RJ01 | Rejection of invention patent application after publication |