CN106126319B - Central processing unit resource allocation methods and calculate node - Google Patents
Central processing unit resource allocation methods and calculate node Download PDFInfo
- Publication number
- CN106126319B CN106126319B CN201610737463.5A CN201610737463A CN106126319B CN 106126319 B CN106126319 B CN 106126319B CN 201610737463 A CN201610737463 A CN 201610737463A CN 106126319 B CN106126319 B CN 106126319B
- Authority
- CN
- China
- Prior art keywords
- resource
- virtual machine
- premium
- pond
- resource pond
- 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.)
- Active
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
- G06F9/46—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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/5011—Pool
Abstract
The present embodiments relate to the resource allocation methods of central processing unit and calculate nodes.Method includes: to be changed according to the quantity of the virtual machine in premium resource pond, dynamically adjusts the quantity allotted of the physical core in the premium resource pond;The quantity allotted of the physical core in the quantity allotted adjustment default resource pond of core is managed according to the premium resource mediocre person;Wherein, the premium resource pond and default resource pond are resource pools obtained from being divided the physical core of central processing unit according to the grade of service of resource pool.According to embodiments of the present invention, by the way that physical core to be divided into different resource pools, and the computing resource of the physical core in different resource pools is isolated from each other between different resource pond, realizes effective distribution of the physical core computing resource for different application scene.
Description
Technical field
The present embodiments relate to virtualization technologies, especially central processing unit resource allocation methods and calculate node.
Background technique
At present in information technology field, virtualization technology is widely used.Virtualization technology provides a kind of energy
Enough abilities by multiple deploying virtual machines on one server, reduce the totle drilling cost of user and improve the reliable of application
Property and flexibility.The management of industry virtual machine central processing unit (Central Processing Unit, abbreviation CPU) is general at present
All over CPU excessive distribution is realized, i.e. the virtual cpu number of the virtual machine number that is greater than physical cpu in server reaches object with this
Reason resource makes full use of.But physical resource anxiety in the case where, how reasonable distribution physical cpu resource, become pendulum exist
One of greatest problem in front.The common priority for being achieved in that setting virtual machine is virtual according to the difference of priority
Machine distributes the cpu resource of different shares.The virtual machine of high priority distributes the physical cpu resource of more share, realizes needle with this
To the user of different demands, different services is provided, is improved service quality.
It is that virtual machine distributes physics cpu resource according to the height of priority, the virtual machine that may cause high priority is in
Monopoly position, and the virtual machine of low priority can not obtain physical cpu resource, be unfavorable for effectively deploying physics cpu resource.
Summary of the invention
The embodiment of the present invention proposes a kind of central processing unit resource allocation methods, by the way that the physical core of CPU to be separated into
Different resource pools realizes that physical core is isolated from each other between different resource pond, and according to the virtual machine of each resource pool service
Quantity, dynamically distribute physical core between different resource pond, realize the effective use of physics nuclear resource, reliably simultaneously
It ensure that the service quality of virtual machine in different resource pond.
In a first aspect, proposing a kind of central processing unit resource allocation methods, which comprises
Changed according to the quantity of the virtual machine in premium resource pond, dynamically adjusts the physical core in the premium resource pond
Quantity allotted;
The distribution number of the physical core in the quantity allotted adjustment default resource pond of core is managed according to the premium resource mediocre person
Amount;
Wherein, the premium resource pond and default resource pond are the grades of service according to resource pool by the object of central processing unit
Resource pool obtained from reason core is divided.
With reference to first aspect, in the first possible implementation, the grade of service includes allowing every virtual machine
Occupy the quantity of physical core computing resource.
With reference to first aspect or the first possible implementation of first aspect, in second of possible implementation
In, the virtual machine in the default resource pond shares the computing resource of whole physical cores in the default resource pond.
With reference to first aspect or first and second kinds of possible implementations of first aspect, in the third possible reality
In existing mode, the premium resource pond includes first order resource pool, wherein allowing every virtual machine in the first order resource pool
Monopolize a physical core computing resource.
With reference to first aspect or first aspect first to the third possible implementation, in the 4th kind of possible reality
In existing mode, the premium resource pond includes second level resource pool, wherein allowing every virtual machine in the second level resource pool
Occupy the 50% of a physical core computing resource.
With reference to first aspect or first to fourth kind of possible implementation of first aspect, in the 5th kind of possible reality
It is described to be changed according to the quantity of the virtual machine in premium resource pond in existing mode, it dynamically adjusts in the premium resource pond
The quantity allotted of physical core, comprising:
When virtual machine quantity in the premium resource pond increases, according to increased virtual machine quantity and described advanced
The grade of service of resource pool determines the physical core for needing the first quantity distributed to the premium resource pond;
The physical core of first quantity is distributed from the default resource pond to the premium resource pond;
Or
When virtual machine quantity in the premium resource pond is reduced, according to the virtual machine quantity of reduction and described advanced
The grade of service of resource pool determines the physical core for needing the second quantity reduced from the premium resource pond;
The physical core of second quantity is distributed from the premium resource pond to the default resource pond.
With reference to first aspect or first to the 5th kind of possible implementation of first aspect, in the 6th kind of possible reality
In existing mode, when the virtual machine quantity in the premium resource pond is reduced, according to the virtual machine quantity of reduction and institute
The grade of service for stating premium resource pond determines the physical core for needing the second quantity reduced from the premium resource pond, from the height
Grade resource pool distributes the physical core of second quantity to the default resource pond, specifically includes: when in the premium resource pond
When not having virtual machine operation, the physical core in the premium resource pond is distributed into the default resource pond.
With reference to first aspect or first to the 6th kind of possible implementation of first aspect, in the 7th kind of possible reality
In existing mode, is changed described according to the quantity of the virtual machine in premium resource pond, dynamically adjusted in the premium resource pond
Physical core quantity allotted before, the method also includes:
Physical core after determining that the virtual machine quantity in the premium resource pond changes, in the premium resource pond
Lazy weight is to meet the grade of service in the premium resource pond.
With reference to first aspect or first to the 6th kind of possible implementation of first aspect, in the 8th kind of possible reality
In existing mode, is changed described according to the quantity of the virtual machine in premium resource pond, dynamically adjusted in the premium resource pond
Physical core quantity allotted before,
The method also includes:
After determining that the virtual machine quantity in the premium resource pond changes, exists in the premium resource pond and refuse to obey
At least one the idle physical core computing resource for the virtual machine being engaged in the premium resource pond.
With reference to first aspect or first to the 8th kind of possible implementation of first aspect, in the 9th kind of possible reality
In existing mode, the method also includes: the object for allowing every virtual machine to occupy for the virtual machine setting in the default resource pond
Manage the upper limit percentage and/or lower percentiles of core computing resource.
With reference to first aspect or first to the 9th kind of possible implementation of first aspect, in the tenth kind of possible reality
In existing mode, the method also includes: every virtual machine, which is set, for the virtual machine in the default resource pond is assessed using physics
The priority of resource is calculated, and is the virtual machine distribution physical core in the default resource pond according to the priority sequence
Computing resource.
Second aspect, proposes a kind of calculate node for virtual system, and the calculate node includes:
Resource adjustment unit dynamically adjusts the height for changing according to the quantity of the virtual machine in premium resource pond
The quantity allotted of physical core in grade resource pool, and the quantity allotted adjustment institute for managing core according to the premium resource mediocre person
State the quantity allotted of the physical core in default resource pond;
The physical core of central processing unit is divided into default for the grade of service according to resource pool by resource allocation unit
Resource pool and premium resource pond.
In conjunction with second aspect, in the first possible implementation, the virtual machine quantity in the premium resource pond
When increase, the resource adjustment unit determines need according to the grade of service of increased virtual machine quantity and the premium resource pond
The physical core for the first quantity to be distributed to the premium resource pond, and from the default resource pond to the premium resource pond
Distribute the physical core of first quantity;
Or
When virtual machine quantity in the premium resource pond is reduced, the resource adjustment unit is according to the virtual machine of reduction
Quantity and the grade of service in the premium resource pond determine the physics for needing the second quantity reduced from the premium resource pond
Core, and distribute from the premium resource pond to the default resource pond physical core of second quantity.
In conjunction with the possible implementation of the first of second aspect or second aspect, in second of possible implementation
In, the calculate node further include:
First determination unit, after determining that the virtual machine quantity in the premium resource pond changes, the height
Physical core lazy weight in grade resource pool is to meet the grade of service in the premium resource pond.
In conjunction with the possible implementation of the first of second aspect or second aspect, in the third possible implementation
In, the calculate node further include: the first determination unit, for determining that the virtual machine quantity in the premium resource pond becomes
There is at least one the idle physical core for the virtual machine that do not serve in the premium resource pond when change, in the premium resource pond
Computing resource.
In conjunction with second aspect or second aspect first to the third possible implementation, in the 4th kind of possible reality
In existing mode, the resource allocation unit is used to allow every virtual machine for every virtual machine setting in the default resource pond
The upper limit percentage and/or lower percentiles of the physical core computing resource of occupancy.
In conjunction with first to fourth kind of possible implementation of second aspect or second aspect, in the 5th kind of possible reality
In existing mode, the resource allocation unit, which is used to set every virtual machine for the virtual machine in the default resource pond, uses physics
The priority of core computing resource, and be the virtual machine distribution physics in the default resource pond according to the priority sequence
The computing resource of core.
According to embodiments of the present invention, by the way that physical core to be divided into different resource pools, and in different resource pools
The computing resource of physical core is isolated from each other between different resource pond, and the physical core realized for different application scene calculates money
Effective distribution in source, while avoiding the occurrence of the monopolization virtual machine for occupying a large amount of computing resources for a long time, it is therefore prevented that physical core calculates money
Source distribution is unbalance, improves the utilization efficiency of the computing resource of physical core.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be in embodiment or description of the prior art
Required attached drawing is briefly described, it should be apparent that, the accompanying drawings in the following description is only some realities of the invention
Example is applied, it for those of ordinary skill in the art, without creative efforts, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is the schematic flow diagram of central processing unit resource allocation methods according to an embodiment of the present invention;
Fig. 2 is the flow chart that virtual machine according to an embodiment of the present invention moves into process specific example;
Fig. 3 is that virtual machine according to an embodiment of the present invention is moved out the flow chart of process specific example;
Fig. 4 is the signaling interaction diagram that virtual machine is migrated from source calculate node to purpose clearing contact;
Fig. 5 is the schematic configuration diagram of calculate node according to an embodiment of the present invention;
Fig. 6 is the schematic configuration diagram of calculate node according to an embodiment of the present invention;
Fig. 7 is another schematic configuration diagram of calculate node according to an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
The technical solution foundation of the embodiment of the present invention can be tied to specified in the virtual cpu (abbreviation VCPU) of virtual machine
The base power of resource pool is realized.It will be appreciated by those skilled in the art that VCPU be tied to resource pool concrete mode and
Implementation method.In order to simple, this is not described in detail herein.
In the description of the embodiment of the present invention, physical core refers to the kernel of physical central processor.Run virtual machine
Calculate node generally comprises multiple CPU to server in other words, and each CPU includes at least one physical core.These physical cores can be with
It is respectively used to different purposes.The physical core that the CPU of resource pool is constituted in the embodiment of the present invention can be the physics of a CPU
Core is also possible to the physical core of multiple CPU;Wherein CPU can be on a server in other words calculate node, can also position
In on multiple servers in other words calculate node, the embodiment of the present invention is not especially limited.
In embodiments of the present invention, a virtual machine may include multiple VCPU, in order to describe simplicity, with a virtual machine
Including being described for a VCPU.In the course of the description, because including a VCPU, institute in embodiment with a virtual machine
" virtual machine and resource pool are bound " can also be generically expressed as with " VCPU and resource pool are bound ".Those skilled in the art understand that
The meaning of " binding " in the text is as equivalent in meaning with as commonly understood in the art.It will be understood by those skilled in the art that " virtual
Machine is run in resource pool " refer to that virtual machine occupies the computing resource of the physical core in the resource pool, and it is not take up other moneys
The computing resource of physical core in the pond of source.In embodiments of the present invention, the computing resource of physical core refers to that the physical core provides
Computing capability.The virtual machine run in resource pool allows the computing resource occupied to can come from an object of the resource pool
Core is managed, can be from multiple physical cores in the resource pool, the embodiment of the present invention is to this and with no restrictions." allow virtual machine
The computing resource of the physical core of occupancy " refers to the object in the resource pool that the virtual machine for allowing a virtual machine to occupy is run
The computing resource for managing core can also indicate in resource pool that the virtual machine for allowing a VCPU of the virtual machine to occupy is run
Physical core computing resource, or can indicate the money that the virtual machine for allowing whole VCPU of the virtual machine to occupy is run
The computing resource of physical core in the pond of source.In the following description, in order to describe clearly, the embodiment of the present invention is with the whole of virtual machine
The computing resource for the physical core in resource pool that the virtual machine that VCPU is occupied is run is as " physical core that virtual machine occupies
The example of computing resource " is described.
Fig. 1 is the schematic flow diagram of central processing unit resource allocation methods according to an embodiment of the present invention.As shown in Figure 1,
Method 100 includes:
110: being changed according to the quantity of the virtual machine in premium resource pond, dynamically adjust the object in the premium resource pond
Manage the quantity allotted of core;
120: the physical core in the default resource pond is adjusted according to the quantity allotted that the premium resource mediocre person manages core
Quantity allotted;
Wherein, the premium resource pond and default resource pond are the grades of service according to resource pool by the object of central processing unit
Resource pool obtained from reason core is divided.
In embodiments of the present invention, each resource pool serves the virtual machine run in the resource pool, the clothes of resource pool
Business grade corresponds to the grade of service required by the virtual machine run in the resource pool." grade of service " described here is for example,
The quantity for the physical core computing resource for allowing every virtual machine to occupy, used in the embodiment of the present invention and latter embodiments
" occupancy " refers to the physical core computing resource that every virtual machine is actually allocated in resource pool.For example, in a premium resource pond
In, allow a virtual machine to occupy the 50% of a physical core computing resource, refers to and be actually allocated in the premium resource pond
The 50% of one physical core computing resource of virtual machine.For example, allowing virtual machine to occupy one in another premium resource pond
Physical core computing resource, the i.e. computing resource to a virtual machine one physical core of practical distribution.
In embodiments of the present invention, logic isolation between different resource pond, i.e., the physical core in each resource pool only service
In the virtual machine run in the resource pool, the virtual machine to run in the resource pool provides physical core computing resource.Here institute
A possibility that physical isolation is not precluded in the logic isolation stated.In a kind of possible implementation according to an embodiment of the present invention
In, different resource pools can be even physically isolated, for example, the physical core in different resource pools is mounted on different calculating sections
Point on.
In embodiments of the present invention, the virtual machine run in default resource pond shares whole physical cores in default resource pond
Computing resource.The set-up mode in premium resource pond can be arranged according to specific application demand.In a kind of possible realization
In mode, premium resource pond includes first order resource pool, wherein allowing every virtual machine exclusive one in the first order resource pool
The computing resource of a physical core, that is to say, that the grade of service of first order resource pool is 100%.It only include one in virtual machine
In the case where VCPU, that is to say, that physics nuclear volume in first order resource pool and the virtual machine that is run in the resource pool
Quantity is equal.First order resource pool can guarantee to the meter in virtual machine one physical core of round-the-clock practical distribution wherein run
Resource is calculated, but " computing resource of a physical core " may be from a physical core or multiple objects in the resource pool
Manage core.In alternatively possible implementation, other than first order resource pool, premium resource pond further includes second level resource
Pond.The virtual machine run in the resource pool of the second level allows to occupy the 50% of a physical core computing resource, that is to say, that second
The grade of service of grade resource pool is 50%.In other words, in the resource pool of the second level, one physics of the practical distribution of every virtual machine is given
The 50% of the computing resource of core, but " the 50% of the computing resource of a physical core " may be from one in the resource pool
A physical core or multiple physical cores.Alternatively possible implementation according to an embodiment of the present invention, premium resource pond can also
Only to include second level resource pool.And alternatively possible implementation according to an embodiment of the present invention, in premium resource pond also
It may include third level resource pool etc., assessed wherein the virtual machine run in third level resource pool allows to occupy a physics
Calculate the 33% of resource.It will be understood by those skilled in the art that can be arranged in premium resource pond according to specific application demand
The grade of service in more or less child resource pond, each child resource pond can be different.
According to embodiments of the present invention, step 110 can specifically include:
When virtual machine quantity in the premium resource pond increases, according to increased virtual machine quantity and described advanced
The grade of service of resource pool determines the physical core for needing the first quantity distributed to the premium resource pond;
The physical core of first quantity is distributed from the default resource pond to the premium resource pond;
Or
When virtual machine quantity in the premium resource pond is reduced, according to the virtual machine quantity of reduction and described advanced
The grade of service of resource pool determines the physical core for needing the second quantity reduced from the premium resource pond;
The physical core of second quantity is distributed from the premium resource pond to the default resource pond.
When not having virtual machine operation in the premium resource pond, the physical core in the premium resource pond is distributed into institute
State default resource pond.It include below default resource pond and premium resource pond with resource pool, wherein premium resource pond includes the first order
It is illustrated for resource pool and second level resource pool.It should be noted that example set forth herein is just to illustrating this hair
The embodiment of bright embodiment, rather than the embodiment of the present invention is limited.When describing the embodiment of the present invention, run on
The virtual machine of the resource pool of the different grades of service, is referred to as the virtual machine of corresponding with service grade, for example, running on the first order
The virtual machine of resource pool is properly termed as first order virtual machine, runs on the virtual machine in default resource pond, is properly termed as default Virtual
Machine.
It is whole due to there is no virtual machine operation in premium resource pond under initial situation, such as after system boot powers on
Physical core distributes in default resource pond.According to the difference of application scenarios, starts virtual machine and be added into different resource pools
Or virtual machine is moved into different resource pools,.Process in contrast to this closes the process of virtual machine or moves virtual machine
Out.For example, it is desired to which first order virtual machine, that is, allow to occupy the virtual machine of physical core computing resource 100%.It then needs in the first order
First order virtual machine is created in resource pool.Those skilled in the art understand that and know the creation and start-up course of virtual machine, therefore
In order to simple, it is not detailed herein.It is empty according to the increased first order when first order resource pool increases first order virtual machine
The quantity of quasi- machine determines the physical core for needing the first quantity distributed to first order resource pool, then distributes from default resource pond
The physical core of first quantity is to first order resource pool.If according to concrete application scene, it is desirable to reduce the number of first order virtual machine
Amount, such as the first order virtual machine of the second quantity moved out from first order resource pool or the first order of the second quantity is virtual
Organ closes.For the first virtual machine of the second quantity to be moved out, while the first order virtual machine of the second quantity is moved out,
Determine the physical core for needing to reduce by the second quantity from first order resource pool.Then, by the second quantity in first order resource pool
Physical core be assigned in default resource pond.It is thus achieved that being changed according to the quantity of the virtual machine in first order resource pool, move
It adjusts to state the quantity allotted of the physical core in the first order resource pool and adjusts the physical core in default resource pond accordingly
Quantity allotted.
For the resource pool of the second level, the mode of adjustment is similar.Difference place is, the in the resource pool of the second level
The 50% of the computing resource of secondary virtual machine one physical core of practical distribution.
According to embodiments of the present invention, in one possible implementation, before step 110, the method also includes:
Physical core after determining that the virtual machine quantity in the premium resource pond changes, in the premium resource pond
Lazy weight is to meet the computational resource requirements in the premium resource pond.For example, in the first order resource pool in premium resource pond
When increasing by 1 first order virtual machine, first order resource pool is unable to satisfy the computing resource of whole virtual machines in first order resource pool
Demand, because every virtual machine is allowed to need the computing resource of an exclusive physical core in first order resource pool.It is also possible to depositing
In following situations, such as in the second level resource pool in premium resource pond, there are 3 second level virtual machines, allow every second level
Virtual machine occupies the 50% of a physical core computing resource, therefore three second level virtual machines need to distribute in the resource pool of the second level
Two physical cores, but after the computing resource for being actually allocated to three second level virtual machine physical cores, an also surplus physical core
The 50% of computing resource.In this case, it if increasing a second level virtual machine in the resource pool of the second level, not will lead to
Second level resource pool is unsatisfactory for the computational resource requirements of whole second level virtual machines.In this way, can will be remained in the resource pool of the second level
The computing resource of a physical core of remaininging 50% distributes to the virtual machine newly increased.In such a case it is not necessary to from default
Resource pool distributes new physical core to second level resource pool.
According to embodiments of the present invention, in one possible implementation, after step 110, the method also includes:
After determining that the virtual machine quantity in the premium resource pond changes, exists in the premium resource pond and refuse to obey
At least one the idle physical core computing resource for the virtual machine being engaged in the premium resource pond.It is described here " idle at least
One physical core computing resource " may be from different physical cores, i.e., the summation of the computing resource from different physical cores is suitable
In the computing resource of a physical core.For example, move out in the first order resource pool in premium resource pond 1 first order virtual machine it
Afterwards, then occurs an idle physical core computing resource in first order resource pool.In this case, which can
It a physical core to be assigned in default resource pond, is used for the virtual machine operated in default resource pond, so as to effectively sharp
With the computing resource of physical core.
Such as after 1 second level virtual machine of moving out in the second level resource pool in premium resource pond, it is possible to not
There is an idle physical core computing resource, because being that every virtual machine is allowed to occupy a physical core in the resource pool of the second level
Computing resource 50%.Therefore, there is at least one idle physical core only in determining second level resource pool and calculate money
In the case where source, physical core can be just allocated back in default resource pond from the second level resource pool.
As previously mentioned, the default Virtual machine in default resource pond shares the calculating money of whole physical cores in default resource pond
Source.According to embodiments of the present invention, shared mode can determine according to specific circumstances.In the first possible implementation
In, default Virtual machine can fifty-fifty share the computing resource of whole physical cores in default resource pond.It is possible according to the first
A kind of improvement project of implementation can set the physics for allowing every default Virtual machine in default resource pond that can occupy
The upper limit percentage of core computing resource, such as allow the occupied physical core computing resource of every virtual machine in default resource pond
Upper limit percentage is the 10% of a physical core computing resource, thus prevents the appearance in default resource pond from monopolizing virtual machine, i.e.,
Occupy the virtual machine of a large amount of computing resources.According to another improvement project, in order to guarantee the meter of virtual machine in default resource pond
Performance is calculated, the lower percentiles of the computing resource for the physical core for allowing every virtual machine that can occupy can be set, such as is defaulted
The lower percentiles of the permission occupied physical core computing resource of every virtual machine in resource pool are that a physical core calculates money
The 5% of source.In the case where determining the upper limit percentage and lower limit of computing resource of physical core that every virtual machine can occupy,
Before implementation steps 110, the method 100 includes:
According in default resource pond virtual machine quantity and the default resource pond in virtual machine allow occupy physics assess
Resource lower percentiles are calculated to determine that default resource pond there are enough resource allocations to give premium resource pond.Alternatively, in default resource
In the case that pond increases virtual machine, determines that the capacity in default resource pond has not yet reached the upper limit, i.e., can also accommodate additional void
Quasi- machine.
For example, it is assumed that the lower percentiles for the computing resource that every virtual machine occupies physical core in permission default resource pond are
5%, and the upper limit percentage for the computing resource that every virtual machine occupies physical core in default resource pond is allowed to be 10%.At this point,
Assuming that default resource pond includes 10 physical cores, then in default resource pond, up to object can be provided with 200 default Virtual machines
Core computing resource is managed, at least can provide physical core computing resource for 100 default Virtual machines.If operating in default resource pond
In virtual machine quantity be 185.If necessary to increase by 1 first order virtual machine, that is, need from default resource pond to advanced money
First order resource pool in the pond of source distributes 1 physical core, and according to calculated result, if distributing an object into first resource pond
After managing core, remaining physics assesses calculation resource and is at most only 180 default Virtual machines offer physical cores calculating in default resource pond
Resource, and 185 default Virtual machines are run in default resource pond now.Then the method 100 includes: that refusal increases advanced void
The request of quasi- machine.In this example, refusal increases first order virtual machine.Certainly, in the embodiment of the present invention in another case, can
With one physical core of first resource pond distribution into premium resource pond, while adjusting every default Virtual machine in default resource pond and accounting for
The lower limit of physical core computing resource, or according to deploying virtual machine strategy in default resource pond, it is empty to close corresponding default
Quasi- machine, to meet the lower limit for the physical core computing resource that existing every default Virtual machine occupies in default resource pond.
For another example assuming to allow the lower limit hundred of the computing resource of every virtual machine one physical core of occupancy in default resource pond
Divide than being 5%, and the upper limit percentage for the computing resource that every virtual machine occupies a physical core in permission default resource pond is
10%.At this time, it is assumed that default resource pond includes 10 physical cores, and the virtual machine quantity operated in default resource pond is 185.
If increasing by 1 virtual machine in default resource pond, that is, increase a default Virtual machine, then according to calculated result, default resource pond
Capacity can also accommodate 3 default Virtual machines, therefore allow into default resource pond increase by 1 virtual machine.
It, can be further preferential for the default Virtual machine setting in default resource pond in alternatively possible implementation
Grade.Physical core computing resource in default resource pond preferentially meets the demand of the high default Virtual machine of priority.Due to setting
Every default Virtual machine is allowed to occupy the upper limit percentage and the upper limit of physical core computing resource, even if the default Virtual of high priority
Machine will not become " monopolization virtual machine " and cause the default Virtual machine of low priority that can not be assigned to computing resource.
One kind of the embodiment of the present invention is completely described with the detailed process example that second level resource pool increases virtual machine below
Realization process.Fig. 2 is the flow chart that virtual machine according to an embodiment of the present invention moves into process specific example.As shown in Fig. 2, second
Grade resource pool increase virtual machine process 200 include:
210: virtual machine 1 is moved into second level resource pool by user's request;
220: system judges whether physical core computing resource is sufficient;
That is, system is according to the virtual machine in current premium resource pond in first order resource pool and second level resource pool
The computing resource lower limit percentage that every virtual machine distributes in virtual machine quantity and default resource pond in quantity, default resource pond
Than, it is determined whether there are enough computational resource allocations to give second level resource pool.In this example, since physical core may move into second
Grade resource pool, system is it needs to be determined that at least one physical core of default resource pond can be assigned to second level resource pool.In physics
In the case that nuclear resource is sufficient, process proceeds to step 230, and in the insufficient situation of physics nuclear resource, process jumps to step
Rapid 270.
230: in the case where physical core computing resource is sufficient, after system judges that virtual machine 1 moves into second level resource pool, being
It is no to need to distribute new physical core for second level resource pool;
In view of allow every virtual machine to occupy a physical core computing resource in the resource pool of the second level 50%, so virtually
Machine 1 is moved into before the resource pool of the second level, if in the case that the virtual machine quantity in the resource pool of the second level is odd number, second level money
The 50% of the computing resource of source pond one physical core of residue not yet distributes.In such a case it is not necessary to be second level resource pool
Distribute new physical core.On the contrary, before virtual machine 1 moves into second level resource pool, if the virtual machine in the resource pool of the second level
In the case that quantity is even number, needs to distribute a physical core from default resource pond to second level resource pool, be provided for virtual machine 1
Computing resource.
240: in the case where not needing to distribute new physical core for second level resource pool, virtual machine 1 being moved into the second level
Resource pool, and the computing resource 50% of a remaining physical core is distributed into the virtual machine 1;
250: in the case where needing to distribute new physical core for second level resource pool, distributing 1 object from default resource pond
It manages core and moves into second level resource pool to second level resource pool, and by virtual machine 1, provide a physical core for the virtual machine 1 and calculate
The 50% of resource;
260: default resource pond accordingly adjusts distribution of the computing resource of physical core between the virtual machine in default resource.
According to foregoing teachings, default resource mediocre person manages the adjustment mode of core computing resource according to the specific of default resource pond
It is arranged different.For example, fifty-fifty sharing the calculating of the physical core in default resource pond in the virtual machine in default resource pond
In the case where resource, due to reducing a physical core, then the practical physical core distributed of each virtual machine in default resource pond
Computing resource is accordingly turned down.
270: virtual machine 1 not being allowed to move into second level resource pool.
In such a case, it is possible to issue the user with notice by message, the physical core computing resource in expression system is not
Foot, does not allow virtual machine 1 to move into.
It is similar with process shown in Fig. 2 that virtual machine moves into first order resource pool, the process in default resource pond.
Corresponding with process shown in Fig. 2, Fig. 3 is the process of virtual machine of moving out from second level resource pool.As shown in figure 3, process
300 include:
310: user's request moves out virtual machine 2 from second level resource pool;
320: system judges that virtual machine 2 is moved out after the resource pool of the second level, if needs to reduce the object in the resource pool of the second level
Manage nuclear volume;
In view of allow every virtual machine to occupy a physical core computer in the resource pool of the second level 50%, so virtual machine
2 move out before the resource pool of the second level, if in the case that the virtual machine quantity in the resource pool of the second level is odd number, second level resource
The 50% of the computing resource of pond one physical core of residue not yet distributes.In this case, virtual machine 2 is moved out second level resource pool
Later, then there is a physical core computing resource idle, need the physical core of second level resource pool reducing 1.On the contrary, virtual
Machine 2 is moved out before the resource pool of the second level, if in the case that the virtual machine quantity in the resource pool of the second level is even number, it cannot be from
The quantity of second level resource pool reduction physical core.
330: in the case where needing to reduce the physics nuclear volume of second level resource pool, i.e. the existing void of second level resource pool
When quasi- machine quantity is odd number, the 50% of the computing resource of second level resource pool one physical core of residue is not yet distributed.This
In the case of, virtual machine 2 is moved out after the resource pool of the second level, then has a physical core computing resource idle, by a physical core point
It is fitted in default resource pond, process jumps to step 350;
340: in the case where not needing to reduce the physics nuclear volume of second level resource pool, virtual machine 2 being moved out second
Grade resource pool;
350: default resource pond accordingly adjusts distribution of the computing resource of physical core between the virtual machine in default resource.
In the embodiment of the present invention, if virtual machine is all moved out in the resource pool of the second level, by the object in Secondary resource pond
Reason core is fully allocated to default resource pond.
In above-described embodiment for including first order resource pool and second level resource pool in premium resource pond, according to reality
Application scenarios, premium resource pond can also no longer include the child resource pond as first order resource pool and second level resource pool.
According to practical application scene, premium resource pond can also not only include first order resource pool and second level resource pool, every height
The grade of service of resource pool can specifically be set according to practical application.
According to foregoing teachings, default resource mediocre person manages core computing resource adjustment being specifically provided with according to default resource pond
Institute is different.For example, fifty-fifty sharing the computing resource of the physical core in default resource pond in the virtual machine in default resource pond
In the case of, due to increasing a physical core, then the computing resource quilt for the physical core that each virtual machine is distributed in default resource pond
It is corresponding to be turned up, to make full use of physical core computing resource.
Fig. 4 is a kind of signaling interaction diagram of concrete application scene according to an embodiment of the present invention.Application scenarios as shown in Figure 4
Virtual machine is described to migrate from source calculate node to purpose calculate node.The side of the embodiment of the present invention as shown in Figure 1,2 and 3
Method can be using in application scenarios shown in Fig. 4.
As shown in figure 4, transition process 400 includes:
410: client issues virtual machine (vm) migration request to cluster controller, and request migrates virtual machine 3 from source calculate node
To purpose calculate node;
420: cluster controller receives the virtual machine (vm) migration that client issues and requests and carry out respective handling;
Cluster controller can execute corresponding processing, such as preempt distribution according to traditional scheduling virtual machine process
Resource.
430: cluster controller issues scheduling request to source calculate node, for virtual machine 3 to be moved out from source calculate node;
440: virtual machine 3 is moved out from source calculate node;
In source calculate node, such as executes virtual machine shown in Fig. 3 and move out process.
450: source calculate node is implemented physical core computing resource dynamic and is adjusted;
460: virtual machine 3 moves into purpose calculate node;
In purpose calculate node, such as executes virtual machine shown in Fig. 2 and move into process.
470: purpose calculate node is implemented physical core computing resource dynamic and is adjusted
480: source calculate node returns to success or failure message of moving out to cluster controller.
490: cluster controller notifies guest virtual machine 3 from source calculate node to the migration results of purpose calculate node.
According to embodiments of the present invention, by the way that physical core to be divided into different resource pools, and in different resource pools
The computing resource of physical core is isolated from each other between different resource pond, and the physical core realized for different application scene calculates money
Effective distribution in source, while avoiding the occurrence of the monopolization virtual machine for occupying a large amount of computing resources, it is therefore prevented that physical core computing resource point
With unbalance, the utilization efficiency of the computing resource of physical core is improved.
The method of the embodiment of the present invention can be applied in different cloud computing business models, such as in internet data
Under the heart (Internet Data Center, abbreviation IDC) business model, it can provide a user in first order resource pool
Virtual machine, so that the computing resource of the full-time exclusive physical core of virtual machine, and computing resource will not be occupied by other users, for visitor
Family provides stable computing capability.For example, business private clound (Service Private Cloud, abbreviation SPC) business model
Under, the virtual machine provided to client has both required superior performance and service stability, it is also desirable to which a degree of cpu resource is multiple
With then the virtual machine in the first order and second level resource pool all meets the application of this scene.For another example on virtual desktop basis
Under framework (Virtual Desktop Infrastructure, abbreviation VDI) business model, the cost performance of customer requirement virtual machine
It is relatively high, resource reusable.In this case, virtual machine has biggish busy ratio, so the virtual machine in default resource pond
It is suitble to this scene.
The embodiment of the present invention also proposed can be using the calculate node of the method for the embodiment of the present invention.As shown in figure 5, meter
Operator node 500 includes:
Resource adjustment unit 510 dynamically adjusts described for being changed according to the quantity of the virtual machine in premium resource pond
The quantity allotted of physical core in premium resource pond, and the quantity allotted adjustment for managing core according to the premium resource mediocre person
The quantity allotted of physical core in the default resource pond;
The physical core of central processing unit is divided into silent by resource allocation unit 520 for the grade of service according to resource pool
Recognize resource pool and premium resource pond.
In the embodiment of the present invention, each resource pool serves the virtual machine run in the resource pool, in appropriate situation
Under, calculate node according to an embodiment of the present invention can have the other technical characteristics in embodiment of the present invention method.It lifts below
Example explanation.
According to embodiments of the present invention, the grade of service includes the number for allowing every virtual machine to occupy physical core computing resource
Amount.
According to embodiments of the present invention, the virtual machine that the resource allocation unit is used to dispatch in the default resource pond is shared
The computing resource of whole physical cores in the default resource pond.
According to embodiments of the present invention, the premium resource pond includes first order resource pool, wherein the first order resource pool
In every virtual machine allow monopolize a physical core computing resource.
According to embodiments of the present invention, the premium resource pond includes second level resource pool, wherein the second level resource pool
In every virtual machine allow occupy a physical core computing resource 50%.
According to embodiments of the present invention, when the virtual machine quantity in the premium resource pond increases, the resource adjustment is single
Member 510 is determined according to increased virtual machine quantity and the grade of service in the premium resource pond to be needed to the premium resource pond
The physical core of first quantity of distribution, and first quantity is distributed from the default resource pond to the premium resource pond
Physical core;
Or
When virtual machine quantity in the premium resource pond is reduced, the resource adjustment unit 510 is according to the void of reduction
Quasi- machine quantity and the grade of service in the premium resource pond determine the second quantity for needing to reduce from the premium resource pond
Physical core, and distribute from the premium resource pond to the default resource pond physical core of second quantity.
According to embodiments of the present invention, when there is no virtual machine operation in the premium resource pond, the resource adjustment unit
For the physical core in the premium resource pond to be distributed to the default resource pond.
As shown in fig. 6, according to embodiments of the present invention, the calculate node further include:
First determination unit 530, it is described after determining that the virtual machine quantity in the premium resource pond changes
Physical core lazy weight in premium resource pond is to meet the grade of service in the premium resource pond.
According to embodiments of the present invention, first determination unit 530 is also used to determine virtual in the premium resource pond
There is the free time for the virtual machine that do not serve in the premium resource pond when machine quantity changes, in the premium resource pond extremely
A few physical core computing resource.
According to embodiments of the present invention, the resource allocation unit 520 is used to dispatch the virtual machine in the default resource pond
It is shared in the default resource pond according to the percentage of the physical core computing resource of preset permission every virtual machine occupancy
The computing resource of whole physical cores.
According to embodiments of the present invention, the resource adjustment unit 510 is used for according to the physical core in the default resource pond
The percentage of quantity and the preset physical core computing resource for allowing every virtual machine to occupy, determines the default resource pond
The quantity for the virtual machine that can be accommodated.
According to embodiments of the present invention, the resource allocation unit 520 is used to dispatch the virtual machine in the default resource pond
Fifty-fifty share the computing resource of whole physical cores in the default resource pond.
According to embodiments of the present invention, the resource allocation unit 520 is used for as every in the default resource pond virtually
The upper limit percentage and/or lower percentiles for the physical core computing resource that machine setting allows every virtual machine to occupy.
According to embodiments of the present invention, the resource allocation unit 520 is used to set for the virtual machine in the default resource pond
Fixed every virtual machine uses the priority of physical core computing resource, and is the default resource according to the priority sequence
The computing resource of virtual machine distribution physical core in pond.
It is another schematic configuration diagram of calculate node according to an embodiment of the present invention shown in Fig. 7.As shown in fig. 7, calculating
Node 700 includes at least one processor 710, such as CPU, at least one port 720, memory 730 and at least one communication
Bus 740.Communication bus 740 is for realizing the connection communication between these devices.Processor 710 is for executing memory 730
The executable module of middle storage, such as computer program;Calculate node optionally includes user interface 750, including but not limited to
Display, keyboard and pointing device, such as mouse, trace ball (trackball), touch-sensitive plate or touch sensitive display screen.Memory
730 may include high speed RAM memory, it is also possible to and it further include nonvolatile memory (non-volatile memory), such as
At least one magnetic disk storage.The communication connection of the calculate node and cluster controller is realized by least one port 720, is led to
Cross the communication connection between at least another port 720 and at least one other calculate node.
In some embodiments, memory 730 stores following element, executable modules or data structures, or
Their subset of person or their superset:
Operating system 732 includes various system programs, for realizing various basic businesses and hardware based of processing
Business;
Application module 734 includes various application programs, for realizing various applied business.
Including but not limited to resource adjustment unit (510), resource allocation unit (520), first determine in application module 734
Unit (530).
The corresponding units of each unit implemented in illustrated embodiment referring to figs. 5 and 6 in application module 732, herein
It does not repeat.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually
It is implemented in hardware or software, the specific application and design constraint depending on technical solution.Professional technician
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
The scope of the present invention.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In several embodiments provided herein, it should be understood that disclosed systems, devices and methods, it can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
It divides, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components
It can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, it is shown or
The mutual coupling, direct-coupling or communication connection discussed can be through some interfaces, the indirect coupling of device or unit
It closes or communicates to connect, can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.
And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic or disk.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (18)
1. a kind of central processing unit resource allocation methods, which is characterized in that the described method includes:
Changed according to the quantity of the virtual machine in premium resource pond, dynamically adjusts point of the physical core in the premium resource pond
With quantity;
The quantity allotted of the physical core in the quantity allotted adjustment default resource pond of core is managed according to the premium resource mediocre person;
Wherein, the premium resource pond and default resource pond are the grades of service according to resource pool by the physical core of central processing unit
Resource pool obtained from being divided;Wherein, the virtual machine run in the premium resource pond occupies the physics of preset ratio
Core computing resource;Virtual machine is run in the default resource pond;
Wherein, described to be changed according to the quantity of the virtual machine in premium resource pond, it dynamically adjusts in the premium resource pond
The quantity allotted of physical core, comprising:
When virtual machine quantity in the premium resource pond increases, according to increased virtual machine quantity and the premium resource
The grade of service in pond determines the physical core for needing the first quantity distributed to the premium resource pond;From the default resource pond to
The physical core of first quantity is distributed in the premium resource pond;
Or
When virtual machine quantity in the premium resource pond is reduced, according to the virtual machine quantity of reduction and the premium resource
The grade of service in pond determines the physical core for needing the second quantity reduced from the premium resource pond;
The physical core of second quantity is distributed from the premium resource pond to the default resource pond.
2. the method as described in claim 1, which is characterized in that the grade of service includes that every virtual machine is allowed to occupy physics
The quantity of core computing resource.
3. method according to claim 1 or 2, which is characterized in that the premium resource pond includes first order resource pool, wherein
Every virtual machine is allowed to monopolize a physical core computing resource in the first order resource pool.
4. method according to claim 1 or 2, which is characterized in that the premium resource pond includes second level resource pool, wherein
Every virtual machine is allowed to occupy the 50% of a physical core computing resource in the second level resource pool.
5. the method as described in claim 1, which is characterized in that the virtual machine quantity in the premium resource pond is reduced
When, it is determined according to the virtual machine quantity of reduction and the grade of service in the premium resource pond and needs to subtract from the premium resource pond
The physical core of the second few quantity distributes the physics of second quantity from the premium resource pond to the default resource pond
Core specifically includes: when not having virtual machine operation in the premium resource pond, the physical core in the premium resource pond being distributed
To the default resource pond.
6. method according to claim 1 or 2, which is characterized in that
Changed described according to the quantity of the virtual machine in premium resource pond, dynamically adjusts the physics in the premium resource pond
Before the quantity allotted of core,
The method also includes:
Physics nuclear volume after determining that the virtual machine quantity in the premium resource pond changes, in the premium resource pond
The grade of service being insufficient in the premium resource pond.
7. method according to claim 1 or 2, which is characterized in that
Changed described according to the quantity of the virtual machine in premium resource pond, dynamically adjusts the physics in the premium resource pond
Before the quantity allotted of core,
The method also includes:
After determining that the virtual machine quantity in the premium resource pond changes, exists in the premium resource pond and do not serve
At least one idle physical core computing resource of virtual machine in the premium resource pond.
8. it is method according to claim 1 or 2, the method also includes:
For the upper limit hundred for the physical core computing resource that the virtual machine setting in the default resource pond allows every virtual machine to occupy
Divide ratio and/or lower percentiles.
9. it is method according to claim 1 or 2, the method also includes:
Every virtual machine is set using the priority of physical core computing resource for the virtual machine in the default resource pond, and according to
The priority sequence is the computing resource of the virtual machine distribution physical core in the default resource pond.
10. a kind of calculate node for virtual system, which is characterized in that the calculate node includes:
Resource adjustment unit dynamically adjusts the advanced money for changing according to the quantity of the virtual machine in premium resource pond
The quantity allotted of physical core in the pond of source is managed in the quantity allotted adjustment default resource pond of core according to the premium resource mediocre person
Physical core quantity allotted;
The physical core of central processing unit is divided into default resource for the grade of service according to resource pool by resource allocation unit
Pond and premium resource pond;Wherein, the physical core that the virtual machine run in the premium resource pond occupies preset ratio calculates money
Source;Virtual machine is run in the default resource pond;
When virtual machine quantity in the premium resource pond increases, the resource adjustment unit is according to increased virtual machine quantity
And the grade of service in the premium resource pond determines the physical core for needing the first quantity distributed to the premium resource pond, and
And the physical core of first quantity is distributed from the default resource pond to the premium resource pond;
Or
When virtual machine quantity in the premium resource pond is reduced, the resource adjustment unit is according to the virtual machine quantity of reduction
And the grade of service in the premium resource pond determines the physical core for needing the second quantity reduced from the premium resource pond, and
And the physical core of second quantity is distributed from the premium resource pond to the default resource pond.
11. calculate node as claimed in claim 10, which is characterized in that
The grade of service includes the quantity for allowing every virtual machine to occupy physical core computing resource.
12. calculate node as described in claim 10 or 11, which is characterized in that
The premium resource pond includes first order resource pool, wherein allowing every virtual machine exclusive one in the first order resource pool
A physical core computing resource.
13. calculate node as described in claim 10 or 11, which is characterized in that
The premium resource pond includes second level resource pool, wherein every virtual machine is allowed to occupy one in the second level resource pool
The 50% of a physical core computing resource.
14. calculate node as described in claim 10 or 11, which is characterized in that
When not having virtual machine operation in the premium resource pond, the resource adjustment unit is used for will be in the premium resource pond
Physical core distribute to the default resource pond.
15. calculate node as described in claim 10 or 11, the calculate node further include:
First determination unit, after determining that the virtual machine quantity in the premium resource pond changes, the advanced money
Physical core lazy weight in the pond of source is to meet the grade of service in the premium resource pond.
16. calculate node as described in claim 10 or 11, the calculate node further include:
First determination unit, when for determining that the virtual machine quantity in the premium resource pond changes, the premium resource
There is at least one the idle physical core computing resource for the virtual machine that do not serve in the premium resource pond in pond.
17. calculate node as described in claim 10 or 11, which is characterized in that
The resource allocation unit is used to allow every virtual machine to occupy for every virtual machine setting in the default resource pond
Physical core computing resource upper limit percentage and/or lower percentiles.
18. calculate node as described in claim 10 or 11, which is characterized in that
The resource allocation unit is used to set every virtual machine for the virtual machine in the default resource pond and be assessed using physics
The priority of resource is calculated, and is the virtual machine distribution physical core in the default resource pond according to the priority sequence
Computing resource.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610737463.5A CN106126319B (en) | 2012-08-31 | 2012-08-31 | Central processing unit resource allocation methods and calculate node |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610737463.5A CN106126319B (en) | 2012-08-31 | 2012-08-31 | Central processing unit resource allocation methods and calculate node |
CN201280002398.1A CN103180830B (en) | 2012-08-31 | 2012-08-31 | Central processing unit resource allocation methods and calculating node |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201280002398.1A Division CN103180830B (en) | 2012-08-31 | 2012-08-31 | Central processing unit resource allocation methods and calculating node |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106126319A CN106126319A (en) | 2016-11-16 |
CN106126319B true CN106126319B (en) | 2019-11-01 |
Family
ID=57280872
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610737463.5A Active CN106126319B (en) | 2012-08-31 | 2012-08-31 | Central processing unit resource allocation methods and calculate node |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106126319B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106776044B (en) * | 2017-01-11 | 2020-02-04 | 深圳鲲云信息科技有限公司 | Hardware acceleration method and system based on data flow |
CN109144706A (en) * | 2017-06-15 | 2019-01-04 | 阿里巴巴集团控股有限公司 | A kind of dynamic allocation method of cpu resource, device and physical machine |
CN110750354B (en) * | 2018-07-24 | 2023-01-10 | 中国移动通信有限公司研究院 | vCPU resource allocation method, device and computer readable storage medium |
CN113315719A (en) * | 2020-02-27 | 2021-08-27 | 阿里巴巴集团控股有限公司 | Traffic scheduling method, device, system and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101706743A (en) * | 2009-12-07 | 2010-05-12 | 北京航空航天大学 | Dispatching method of virtual machine under multi-core environment |
CN102014159A (en) * | 2010-11-29 | 2011-04-13 | 华中科技大学 | Layered resource reservation system under cloud computing environment |
CN102176723A (en) * | 2011-03-25 | 2011-09-07 | 北京航空航天大学 | Manufacturing cloud system for supporting on-demand use and dynamic collaboration of manufacturing resources and manufacturing capacities |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7362705B2 (en) * | 2004-05-13 | 2008-04-22 | International Business Machines Corporation | Dynamic load-based credit distribution |
US7673113B2 (en) * | 2006-12-29 | 2010-03-02 | Intel Corporation | Method for dynamic load balancing on partitioned systems |
US8862738B2 (en) * | 2010-10-18 | 2014-10-14 | International Business Machines Corporation | Reallocating resource capacity among resource pools in a cloud computing environment |
-
2012
- 2012-08-31 CN CN201610737463.5A patent/CN106126319B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101706743A (en) * | 2009-12-07 | 2010-05-12 | 北京航空航天大学 | Dispatching method of virtual machine under multi-core environment |
CN102014159A (en) * | 2010-11-29 | 2011-04-13 | 华中科技大学 | Layered resource reservation system under cloud computing environment |
CN102176723A (en) * | 2011-03-25 | 2011-09-07 | 北京航空航天大学 | Manufacturing cloud system for supporting on-demand use and dynamic collaboration of manufacturing resources and manufacturing capacities |
Also Published As
Publication number | Publication date |
---|---|
CN106126319A (en) | 2016-11-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103180830B (en) | Central processing unit resource allocation methods and calculating node | |
US10623481B2 (en) | Balancing resources in distributed computing environments | |
CN105052074B (en) | For providing method, system and the computer-readable medium of virtualization diameter network framework and the diameter resource instances for portfolio to be routed to D ynamic instantiation | |
JP5510556B2 (en) | Method and system for managing virtual machine storage space and physical hosts | |
US9304803B2 (en) | Cooperative application workload scheduling for a consolidated virtual environment | |
KR100956636B1 (en) | System and method for service level management in virtualized server environment | |
CN111966500A (en) | Resource scheduling method and device, electronic equipment and storage medium | |
CN108337109A (en) | A kind of resource allocation methods and device and resource allocation system | |
CN103729236B (en) | A kind of method for limiting cloud computing user resources using amount | |
CN103051564A (en) | Dynamic resource allocation method and device | |
CN103455378A (en) | Resource distribution method of virtual clusters, device and system | |
US11740921B2 (en) | Coordinated container scheduling for improved resource allocation in virtual computing environment | |
CN103516623A (en) | Resource distribution method and system | |
CN112269641A (en) | Scheduling method, scheduling device, electronic equipment and storage medium | |
CN106126319B (en) | Central processing unit resource allocation methods and calculate node | |
CN106909462A (en) | A kind of cloud resource regulating method and device | |
CN111104219A (en) | Binding method, device, equipment and storage medium of virtual core and physical core | |
CN114996003A (en) | Cloud service deployment method and device, electronic equipment and storage medium | |
CN114489922A (en) | Virtual instance setting method and device | |
CN107562510B (en) | Management method and management equipment for application instances | |
CN110750330A (en) | Virtual machine creating method, system, electronic equipment and storage medium | |
KR20190061241A (en) | Mesos process apparatus for unified management of resource and method for the same | |
TWI739235B (en) | A gpu computing resource assignment system and method thereof | |
Che et al. | CuteCloud: putting" Credit Union" cloud computing into practice | |
Vondra et al. | Maximizing utilization in private iaas clouds with heterogenous load |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right |
Effective date of registration: 20220223 Address after: 550025 Huawei cloud data center, jiaoxinggong Road, Qianzhong Avenue, Gui'an New District, Guiyang City, Guizhou Province Patentee after: Huawei Cloud Computing Technology Co.,Ltd. Address before: 518129 Bantian HUAWEI headquarters office building, Longgang District, Guangdong, Shenzhen Patentee before: HUAWEI TECHNOLOGIES Co.,Ltd. |
|
TR01 | Transfer of patent right |