CN109144658A - Load-balancing method, device and the electronic equipment of limited resources - Google Patents
Load-balancing method, device and the electronic equipment of limited resources Download PDFInfo
- Publication number
- CN109144658A CN109144658A CN201710499500.8A CN201710499500A CN109144658A CN 109144658 A CN109144658 A CN 109144658A CN 201710499500 A CN201710499500 A CN 201710499500A CN 109144658 A CN109144658 A CN 109144658A
- Authority
- CN
- China
- Prior art keywords
- virtual cpu
- loadtype
- physical machine
- utilization
- statistics
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5061—Partitioning or combining of resources
- G06F9/5066—Algorithms for mapping a plurality of inter-dependent sub-tasks onto a plurality of physical CPUs
-
- 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]
- G06F9/5083—Techniques for rebalancing the load in a distributed system
- G06F9/5088—Techniques for rebalancing the load in a distributed system involving task migration
-
- 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
-
- 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
- G06F2209/5022—Workload threshold
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 embodiment of the present invention provides the load-balancing method, device and electronic equipment of a kind of limited resources, wherein method includes: the utilization of resources data of each virtual cpu in the physical machine for obtain affiliated same cloud unit;The cloud unit is by multiple physics units at being virtualized out multiple virtual cpus in each physical machine, a cloud unit is used to provide Service Source to specified application cluster;Data statistics is carried out by preset loadtype to the utilization of resources data of each virtual cpu, to obtain corresponding loading statistics under different loads type;If the loading statistics meet the load balance process condition under respective load type, the resource allocation being adapted with the loadtype is carried out to the corresponding Service Source of the loading statistics.The scheme of the embodiment of the present invention can realize load balancing to limited cloud service resource.
Description
Technical field
This application involves field of computer technology more particularly to a kind of load-balancing methods of limited resources, device and electricity
Sub- equipment.
Background technique
Cloud Server (Elastic Compute Service, ECS) be a kind of processing capacity can elastic telescopic calculating clothes
Business, way to manage is more simple and efficient than physical server, biggest advantage be can dynamic adaptation computing resource, flat
Platform total capacity can receive, application software can be carried out according to the variation of hardware resource can be effective in the case where optimizing accordingly
Solve the processing speed of peak time.
Big to promote upper cloud, multiple cloud units can mostly be presented in cloud, and each cloud unit corresponding with service one is relatively complete
Business group, that is to say, that cloud unit be it is closed, cloud unit can complete independent offer is big promote during every business
Service.During resource allocation deployment, it will be matched in conjunction with service traffics, and resource allocation is carried out based on the globality of unit
Deployment.After first resource bid sequence, distribution, volume of business initialization is completed, and starts verifying business correctness, subsequently into
Full link pressure test is received.During pressing survey, there are physical machines to load performance unevenness, and application obscure portions node loads higher show
As.It is balanced in order to solve service node overall load, it needs to carry out effective resource and transfers to other use, realize load balancing.With routine
Load balancing implement difference: cloud element resources pond be often prior budget and specify this unit carrying flow, this
A is the target and restrictive condition for realizing load balancing.Based on this restrictive condition, load balancing, which needs to concentrate, reduces " overheat "
The ratio of node reduces the maximum value of " hot spot " load in other words.
The defect of the prior art:
Be directed to the load balancing scheme of cloud element resources in the prior art: high load node executes offline processing, then weighs
New application resource.This method is simply direct, it can however not ensuring that the internal load of physical machine tends to balanced;In addition, high load
The load of node persistent anomaly, there is a problem inside reaction business to a certain degree, if offline processing, cannot retain example,
And then it cannot effectively check problem.Belong to the scheme damaged by the way of current limliting, although application load can be limited certain
Range, but the node utilization rate of low-load cannot be promoted.
In addition, cloud element resources initialization after, in resource pool surplus resources be limited and in physical machine
Deploy certain embodiments, it is desirable to the probability very little of large area dilatation, because being all by estimating one when initialization capacity application
It walks in place.Large-scale resource is transferred to other use, and moving costs is higher, and new change can also be introduced after migrating by transferring to other use, this needs full chain again
Road pressure, which is surveyed, tests, and introduces higher cost.
Summary of the invention
The present invention provides a kind of load-balancing method of limited resources, device and electronic equipments, to limited service
Resource realizes load balancing.
In order to achieve the above objectives, the embodiment of the present invention adopts the following technical scheme that
In a first aspect, providing a kind of load-balancing method of limited resources, comprising:
The utilization of resources data of each virtual cpu in the physical machine of same cloud unit belonging to obtaining;The cloud unit is by more
A physics unit is at being virtualized out multiple virtual cpus, a cloud unit is used for specified in each physical machine
Application cluster provides Service Source;
Data statistics is carried out by preset loadtype to the utilization of resources data of each virtual cpu, to obtain difference
Corresponding loading statistics under loadtype;
If the loading statistics meet the load balance process condition under respective load type, unite to the load
It counts corresponding Service Source and carries out the resource allocation being adapted with the loadtype.
Second aspect provides the load-balancing method of another limited resources, comprising:
Obtain the utilization of resources data of at least one virtual cpu on target physical machine;It is empty in each physical machine
It is quasi- to dissolve multiple virtual cpus, for providing Service Source to specified application cluster;
Data statistics is carried out by loadtype to the utilization of resources data of at least one virtual cpu, to obtain difference
Corresponding loading statistics under loadtype;
The corresponding Service Source of the loading statistics under same loadtype mutually fit with the loadtype
The resource allocation answered.
The third aspect provides a kind of load balancing apparatus of limited resources, comprising:
First resource data acquisition module, the money of each virtual cpu in physical machine for obtaining affiliated same cloud unit
Source utilizes data;The cloud unit by multiple physics units at, be virtualized out multiple virtual cpus in each physical machine,
One cloud unit is used to provide Service Source to specified application cluster;
Second load data statistical module presses preset load class for the utilization of resources data to each virtual cpu
Type carries out data statistics, to obtain corresponding loading statistics under different loads type;
Information resources scheduling module, if meeting the load balancing under respective load type for the loading statistics
Treatment conditions then carry out the resource allocation being adapted with the loadtype to the corresponding Service Source of the loading statistics.
Fourth aspect provides the load balancing apparatus of another limited resources, comprising:
Secondary resource data acquisition module, for obtaining the utilization of resources of at least one virtual cpu on target physical machine
Data;Multiple virtual cpus are virtualized out in each physical machine, for providing Service Source to specified application cluster;
Second load data statistical module, for the utilization of resources data at least one virtual cpu by load class
Type carries out data statistics, to obtain corresponding loading statistics under different loads type;
Secondary resource scheduling module, for the corresponding Service Source of the loading statistics under same loadtype
Carry out the resource allocation being adapted with the loadtype.
5th aspect, provides a kind of electronic equipment, comprising:
Memory, for storing program;
Processor is coupled to the memory, for executing described program, to be used for:
The utilization of resources data of each virtual cpu in the physical machine of same cloud unit belonging to obtaining;The cloud unit is by more
A physics unit is at being virtualized out multiple virtual cpus, a cloud unit is used for specified in each physical machine
Application cluster provides Service Source;
Data statistics is carried out by preset loadtype to the utilization of resources data of each virtual cpu, to obtain difference
Corresponding loading statistics under loadtype;
If the loading statistics meet the load balance process condition under respective load type, unite to the load
It counts corresponding Service Source and carries out the resource allocation being adapted with the loadtype.
6th aspect, provides another electronic equipment, comprising:
Memory, for storing program;
Processor is coupled to the memory, for executing described program, to be used for:
Obtain the utilization of resources data of at least one virtual cpu on target physical machine;It is empty in each physical machine
It is quasi- to dissolve multiple virtual cpus, for providing Service Source to specified application cluster;
Data statistics is carried out by loadtype to the utilization of resources data of at least one virtual cpu, to obtain difference
Corresponding loading statistics under loadtype;
The corresponding Service Source of the loading statistics under same loadtype mutually fit with the loadtype
The resource allocation answered.
Load-balancing method, device and the electronic equipment of limited resources provided by the invention, by the void in physical machine
The utilization of resources data of quasi- CPU carry out data statistics by different loadtypes, to obtain corresponding under different loads type
Then loading statistics carry out and the loadtype the corresponding Service Source of loading statistics under same loadtype
Adaptable resource allocation, to realize the load balancing of limited cloud service resource.
Above description is only the general introduction of technical scheme, in order to better understand the technological means of the application,
And it can be implemented in accordance with the contents of the specification, and in order to allow above and other objects, features and advantages of the application can
It is clearer and more comprehensible, below the special specific embodiment for lifting the application.
Detailed description of the invention
By reading the following detailed description of the preferred embodiment, various other advantages and benefits are common for this field
Technical staff will become clear.The drawings are only for the purpose of illustrating a preferred embodiment, and is not considered as to the application
Limitation.And throughout the drawings, the same reference numbers will be used to refer to the same parts.In the accompanying drawings:
Fig. 1 is the logical schematic of the load balancing of the limited resources of the embodiment of the present invention;
Fig. 2 is the system construction drawing of the load balancing of the limited resources of the embodiment of the present invention;
Fig. 3 a is the load-balancing method flow chart one of the limited resources of the embodiment of the present invention;
Fig. 3 b is the load-balancing method flowchart 2 of the limited resources of the embodiment of the present invention;
Fig. 4 a is the load balancing apparatus structure chart one of the limited resources of the embodiment of the present invention;
Fig. 4 b is the load balancing apparatus structure chart two of the limited resources of the embodiment of the present invention;
Fig. 5 is the structural schematic diagram one of the electronic equipment of the embodiment of the present invention;
Fig. 6 is the structural schematic diagram two of the electronic equipment of the embodiment of the present invention.
Specific embodiment
Exemplary embodiments of the present disclosure are described in more detail below with reference to accompanying drawings.Although showing the disclosure in attached drawing
Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here
It is limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure
It is fully disclosed to those skilled in the art.
The present invention changes in the prior art, in the load balancing scheme of cloud element resources, only blindness to height
Load node executes offline processing, and applies for resource again, and core concept is, for the void in each physical machine on cloud unit
The utilization of resources data of quasi- CPU carry out data statistics by different loadtypes, then by the loading statistics of statistics and phase
Threshold value is answered to be compared to determine whether Service Source meets the load balance process condition under corresponding loadtype, if full
Foot then shows that the Service Source needs to carry out to handle with the resource allocation that the loadtype is adapted, thus targetedly to not
The case where load imbalance of same type, carries out the resource allocation processing of different modes, realizes that the load of limited service resource is equal
Weighing apparatus.
As shown in Figure 1, the logical schematic of the load balancing for the limited resources of the embodiment of the present invention.In the logic chart
In, for differentiating that the case where Service Source is with the presence or absence of load imbalance and to load imbalance carries out the basic of resource allocation
Foundation is that the utilization of resources data of the virtual cpu on each cloud unit in physical machine (are able to reflect the number of resource utilization
According to), such as utilization rate, the utilization rate of load (load) and input/output (I/O) of virtual cpu.The resource benefit of virtual cpu
It not only can reflect the resource utilization of each virtual cpu with data, but also pressed by the utilization of resources data to virtual cpu
Different levels or angle are counted, and corresponding loading statistics under different loads type can also be obtained.Described is negative
It carries type and refers to the level or angle for being able to reflect the corresponding loading condition of Service Source, such as the Service Source pair to cloud unit
The loading condition answered, can be from different levels such as physical machine, application, single virtual CPU, i.e. different loads type is retouched
It states.
It should be noted that physical machine, application, the relationship between single virtual CPU in resource division are as follows:
One cloud unit is used to provide Service Source to specified application cluster.One cloud unit is by multiple physical mechanisms
At.The cloud unit is the ECS virtual machine virtually dissolved from physical machine, is also considered as providing for stationary applications cluster
The cloud host of cloud service, the present invention in physical machine can be equal with ECS virtual machine, a physical machine can be virtualized into one
ECS virtual machine on this ECS virtual machine, then carries out containerization virtual machine distribution, i.e., this ECS virtual machine is regarded as
One physical machine.Physical machine can also be considered as the carrying entity of an ECS virtual machine in the present invention, and such as multiple physical machines are virtual
Multiple virtual machines are turned to, and these virtual machines are used to provide Service Source to specified application cluster, to constitute an ECS void
Quasi- machine).Multiple examples, the corresponding specific business tine of each example are set in each physical machine.Physical machine has phy chip
Computing unit.For example, the 32 core CPU that we usually say can actually be 16 physical computing units in a physical machine,
After virtualizing by Hyper-Threading, become 32 cores.This 32 core can also regard 32 virtual cpus as.Or referred to as VCPU.One
A application may include multiple examples, and an example can occupy n (n is greater than 0 integer) a virtual cpu.Such as to 32 core CPU into
Row number 0 to 31, it is 0,1,2,3 that an example, which can occupy wherein number, the virtual cpu of this 4 numbers.Multiple virtual cpus pair
Ying Yuyi physical computing unit, for example, a kind of virtual cpu coding rule is: 0,1,2 ... 31, wherein possible 0,15 is corresponding same
One physical computing unit, 1, the 16 corresponding same physical computing unit, and so on 15, the 31 corresponding same physical computing lists
Member.In other words, after 15,31 are virtualized by a physical computing unit, become two virtual cpus, number definition 15,31.
Based on above content, as soon as after the utilization of resources data for obtaining each virtual cpu in the physical machine on cloud unit,
Its load can be determined respectively according to physical machine, application, single virtual CPU these three levels to the Service Source on the cloud unit
Situation.
In logical schematic shown in Fig. 1, by the utilization of resources data of each virtual cpu in the physical machine on cloud unit
It is counted by different loadtypes, to obtain physical machine load data, application cluster load data and virtual cpu load
Then the loading statistics of these three levels of data carry out load data calculating to these three loading statistics, thus
To physical machine high load node, using high load node, virtual cpu high load node and corresponding load value.Finally, according to
Each load value carries out classification calculating to the load imbalance problem of current cloud unit Service Source, and available three kinds of loads are uneven
The case where weighing apparatus:
1, resource allocation is uneven, refers to the load imbalance of physical machine level, i.e., embodies on the whole between each physical machine
Example on physical machine high load node can be pinpointed move to other physical machine low-load nodes at this time by load imbalance
On, to reach the load balancing between each physical machine.
2, using off-capacity, refers to the load imbalance of application, i.e., embodied on the whole between the example of each application
Resource on application high load node can be carried out dilatation at this time, to meet this using resource by load imbalance out
Demand, to reach the load balancing between each application.
3, business self problem refers to the load imbalance of the occupied virtual cpu level of each example inside each application, and
And it is this it is unbalanced be embodied in certain examples occupied virtual cpus always duration show high load, and in resource
Original allocation when, it is sufficient for keeping for the resource of the example.In this case it is likely to business itself (code level)
It goes wrong.At this point it is possible to close the corresponding service of the business, to carry out defect investigation, fundamentally solve shared by the example
The case where virtual cpu high load, realizes the load balancing between each virtual cpu.
Finally, after determining the situation of load imbalance, so that it may according to above-mentioned money corresponding with each loadtype enumerating
Source programs carry out resource allocation processing to the object (physical machine, application, virtual cpu) for needing to carry out resource allocation.According to
Implementing result further follows up, and locks to stable Service Source node, is not involved in resource allocation next time.
The logical schematic of load balancing based on limited resources shown in FIG. 1, the embodiment of the invention provides one kind to have
The SiteServer LBS for limiting resource, to calculate loading statistics of the cloud unit under different loads type, and according to load
Statistical data obtains the loadtype of corresponding high load node, carries out reasonable resource allocation according to loadtype, thus real
The load balancing of existing cloud service resource.As shown in Fig. 2, the system includes: the load balancing apparatus of cloud 210 and limited resources
220;Wherein, as shown in Figure 2, the load balancing apparatus 220 of limited resources can be provided separately with cloud 210, also may include
Beyond the clouds in 210, but it is provided separately between each cloud unit in cloud.The load balancing apparatus 220 of limited resources can be with
It is the dispatching platform being arranged in a network, during it can collect full link pressure survey, the letter such as resource, data on each cloud unit
Breath, and resource can be executed to the node for needing to be implemented load balancing and the operations of load balancing such as be transferred to other use.
Cloud 210 includes:
(cloud unit 1 ..., cloud unit n) may include multiple physical machine (physical machines in each cloud unit for multiple cloud units
1 ..., physical machine n), each physical machine can virtually dissolve multiple virtual cpus (virtual cpu 1 ..., virtual cpu n).One cloud list
Member can be used for providing Service Source to specified application cluster.
Wherein, the so-called unit in the present embodiment refers to comprising such as transaction is browsed, placed an order, paying, logistics is with respect to closed loop
Business group, that is to say, that unit includes complete: database, middleware, srvice instance.When these units operate in cloud
In the physical machine of offer, then referred to as cloud unit, is not the physical machine that cloud provides, then referred to as non-cloud unit;For example, Taobao, day cat
Equal shoppings online platform mostly contains by querying commodity information, places an order, pays, the industry of many service constitutions of browsing etc. of trade
Be engaged in group, run required for these business groups or occupy on cloud the various resources of physical machine, including database, various middlewares with
And srvice instance etc. can uniformly be summarised as a cloud unit, i.e., can carry out on each cloud unit similar Taobao,
The complete business group of the platforms such as its cat.Certainly, as the division boundary of business group is different, business group pointed by a cloud unit
Specific business tine it is also different.
The load balancing apparatus 220 of limited resources includes:
Resource data obtains module, the resource benefit of each virtual cpu in physical machine for obtaining affiliated same cloud unit
Use data;The cloud unit is by multiple physics units at being virtualized out multiple virtual cpus, a cloud list in each physical machine
Member is for providing Service Source to specified application cluster;
Wherein, the utilization of resources data of each virtual cpu are to be able to reflect the data of resources of virtual machine utilization power, such as virtual
The utilization rate of CPU, load (load) and the utilization rate of input/output (I/O) etc..
Load data statistical module is counted for the utilization of resources data to each virtual cpu by preset loadtype
According to statistics, to obtain corresponding loading statistics under different loads type;
Wherein, the loadtype refers to the level or angle for being able to reflect the corresponding loading condition of Service Source, example
It, can be from different levels such as physical machine, application, single virtual CPU, i.e., such as to the corresponding loading condition of the Service Source of cloud unit
Different loads types is described.
Resource allocation module, if meeting the load balance process item under respective load type for loading statistics
Part then carries out the resource allocation being adapted with the loadtype to the corresponding Service Source of the loading statistics.
Each loadtype corresponds to a load balance process condition, which is united for measuring
Whether the resource object of meter needs to be performed resource allocation, to achieve the purpose that load balancing.Specifically, if load statistics number
According to the load balance process condition met under respective load type, then illustrate that the resource object has load unevenness, specifically
It is that there are high load nodes, at this time, it may be necessary to mutually fit with the loadtype to the corresponding Service Source of the loading statistics
The resource allocation answered.
Further, the utilization of resources data of above-mentioned each virtual cpu are specifically as follows the utilization rate (virtual cpu of virtual cpu
Utilization rate be much larger than the utilization rate of load and I/O, both hereafter because can be ignored when actually calculating).
Further, above-mentioned loadtype can include: physical machine loadtype, application cluster loadtype and business defect
At least one of loadtype.
Wherein, physical machine loadtype, application cluster loadtype and business defect loadtype be followed successively by with physical machine,
The loading condition of Service Source is described using, these three levels of virtual cpu.
On this basis, from physical machine loadtype, three angles of application cluster loadtype and business defect loadtype
Degree sets out, and further can carry out corresponding module refinement to above-mentioned load data statistical module and resource allocation module.
Load data statistical module can further comprise:
Physical machine load data statistic unit, for by the utilization of resources data of each virtual cpu as unit of physical machine into
Row statistics, to obtain the resource utilization of the virtual cpu in each physical machine, and the result data of statistics is born as physical machine
Carry the loading statistics under type.For example, the utilization rate of each virtual cpu is specifically how many in each physical machine.
Correspondingly, load balance process condition under physical machine loadtype can be with are as follows: the benefit of each virtual cpu in physical machine
First threshold is all larger than with rate.The first threshold is empirical value, such as 35%.In fact, when all virtual cpus in a certain physical machine
Utilization rate when being both greater than 35%, illustrate the physical machine be assigned on the whole itself resource it is just insufficient, need whole to object
The Service Source of reason machine carries out dilatation.
Correspondingly, resource allocation module can further comprise:
Physical machine resource allocation unit, for being all larger than first threshold for the utilization rate of each virtual cpu in physical machine
Physical machine, carries out whole resource capacity expansion and/or resource isolation, and/or by the example of the partial virtual CPU occupied in the physical machine
It moves on the virtual cpu in other physical machines.
Have to the mode of physical machine realization dilatation following several: dilatation can be carried out to the resource in entire physical machine, from
And increase the resource of the occupied virtual cpu of running example in the physical machine;It can also be to the example institute run in the physical machine
The resource of occupancy carries out resource isolation, to not be further added by the load capacity of the physical machine;It can also will occupy in the physical machine
On virtual cpu in the instance migration of partial virtual CPU to other physical machines, has example to void to increase in the physical machine
The usage amount of the resource of quasi- CPU.
Further, in the instance migration to other physical machines for the partial virtual CPU that will be occupied in the physical machine
Resource allocation scheme on virtual cpu, physical machine resource allocation unit, it may also be used in statistics cloud unit, the utilization of virtual cpu
The smallest virtual cpu of rate, and most by the example of the partial virtual CPU in the above-mentioned occupancy physical machine and the utilization rate of virtual cpu
The example that occupies on small virtual cpu carries out location swap, to realize local, point-to-point resource is transferred to other use, to guarantee cloud
The stability of unit on the whole is consistent, and local migration is also change after all, and migration repeatedly is also unreasonable.
Load data statistical module can further comprise:
Application cluster load data statistic unit, the object for servicing the utilization of resources data of each virtual cpu are answered
It is counted as unit of, to obtain the included example of each application in the resource utilization of its occupied virtual cpu, and
Using the result data of statistics as the loading statistics under the application cluster loadtype.For example, each application is included
Example, be how many in the utilization rate of the virtual cpu accordingly occupied.
Correspondingly, load balance process condition under application cluster loadtype can be with are as follows:
Using the example for being included in the utilization rate of its virtual cpu occupied, more than the example of fixed proportion, occupy
The utilization rate of virtual cpu be all larger than second threshold.For example, in an included example of application, the example more than 10%,
The utilization rate for the virtual cpu that should be occupied has been more than second threshold, and such as 45%.In fact, the example included when a certain application
In, there is the example of the ratio greater than 10%, when the utilization rate of the virtual cpu occupied is both greater than 45%, illustrates this using whole
The resource that itself is assigned on body is just insufficient, and the whole Service Source to the application is needed to carry out dilatation.
Correspondingly, resource allocation module can further comprise:
Application cluster resource allocation unit, for being directed to the included example of application in the utilization of its virtual cpu occupied
In rate, more than the example of fixed proportion, the utilization rate of the virtual cpu occupied is all larger than the application of second threshold, by the application
Occupied resource carries out whole resource capacity expansion.
Load data statistical module can further comprise:
Business defect load data statistic unit, the object for servicing the utilization of resources data of each virtual cpu are answered
It is counted as unit of, to obtain the included example of each application in the resource utilization of its occupied virtual cpu, and
Using the result data of statistics as the loading statistics under business defect loadtype.For example, the reality that each application is included
Example, is how many in the utilization rate of the virtual cpu accordingly occupied.
Correspondingly, load balance process condition under business defect loadtype can be with are as follows:
Using the example for being included in the utilization rate of its virtual cpu occupied, the occupied virtual cpu of same instance
Utilization rate is persistently greater than third threshold value.For example, there are fixed certain node instances in an included example of application,
The utilization rate of the virtual cpu of occupancy can not have always been high any more, and be continued above third threshold value, and such as 80%.In fact, when a certain
Using some example for being included, the utilization rate of the virtual cpu occupied is in the state greater than 80% for a long time, illustrates the reality
There may be business defects for example itself, such as there is defect from business code layer face, thus lead to its practical virtual cpu occupied
Utilization rate utilization rate is much bigger than expected.
Correspondingly, resource allocation module can further comprise:
Business defect resource allocation unit, it is described virtual for being occupied for the included example of the application at it
In the utilization rate of CPU, the utilization rate of the occupied virtual cpu of same instance is persistently greater than the application of third threshold value, determines that this is answered
There are business defects for the example in, and close the corresponding service of the example, to carry out defect investigation.
It when determining that business defect occurs in example itself, needs to close the corresponding service of the example, carries out defect investigation, it is excellent
Change example, fundamentally to solve the problems, such as high load.
The SiteServer LBS of limited resources provided in an embodiment of the present invention, by each in the physical machine on cloud unit
The utilization of resources data of virtual cpu carry out data statistics by different loadtypes, to obtain corresponding under different loads type
Loading statistics, then to loading statistics carry out respective load type under load balance process condition criterion, such as
Fruit meets load balance process condition, then be adapted with the loadtype to the corresponding Service Source of the loading statistics
Resource allocation, to realize the load balancing of limited cloud service resource.
The technical solution of the application is further illustrated below by multiple embodiments.
Embodiment one
Based on the scheme thought of the above-mentioned load balancing for carrying out limited resources by loadtype, as shown in Figure 3a, for this
The load-balancing method flow chart one of limited resources shown in inventive embodiments, the execution of this method is main to be had to be shown in Fig. 2
Limit the load balancing apparatus of resource.As shown in Figure 3a, the load-balancing method of the limited resources includes the following steps:
S310 obtains the utilization of resources data of each virtual cpu in the physical machine of affiliated same cloud unit;The cloud unit
By multiple physics units at being virtualized out multiple virtual cpus in each physical machine, a cloud unit is used for specified application
Cluster provides Service Source.
Specifically, it is supervised by utilization of resources data of the monitoring device to each virtual cpu in the physical machine of cloud unit
It surveys and obtains.Wherein, the utilization of resources data of the virtual cpu are to be able to reflect the data of resource utilization, such as virtual
The utilization rate of CPU, load (load) and the utilization rate of input/output (I/O) etc..
S320 carries out data statistics by preset loadtype to the utilization of resources data of each virtual cpu, to obtain difference
Corresponding loading statistics under loadtype;
The utilization of resources data of virtual cpu not only can reflect the resource utilization of each virtual cpu, and by pair
The utilization of resources data of virtual cpu are counted by different level or angle, can also be obtained corresponding under different loads type
Loading statistics.The loadtype refers to the level or angle for being able to reflect the corresponding loading condition of Service Source,
Such as to the corresponding loading condition of the Service Source of cloud unit, can from different levels such as physical machine, application, single virtual CPU,
That is different loads type is described.
The reflection that corresponding loading statistics can quantify under different loads type is negative come what is described with the loadtype
Carry situation.Here loading statistics are the number counted for the utilization of resources data of each virtual cpu by loadtype
According to the specific method and form of statistics are without limitation.
S330, if loading statistics meet the load balance process condition under respective load type, to the load
The corresponding Service Source of statistical data carries out the resource allocation being adapted with the loadtype.
Each loadtype corresponds to a load balance process condition, which is united for measuring
Whether the resource object of meter needs to be performed resource allocation, to achieve the purpose that load balancing.Specifically, if load statistics number
According to the load balance process condition met under respective load type, then illustrate that the resource object has load unevenness, specifically
It is that there are high load nodes, at this time, it may be necessary to mutually fit with the loadtype to the corresponding Service Source of the loading statistics
The resource allocation answered.
What needs to be explained here is that the setting about load balance process condition, and the corresponding tool for carrying out resource allocation
Body embodiment is that without limitation, those skilled in the art can be with reference to the load class enumerated in foregoing teachings in this programme
The specific aspect of type, to set at the content of more diversified loadtype, and load balancing corresponding with the loadtype
Manage bar part and resource allocation scheme.No matter it is to be noted that art technology setting concrete scheme, in we
Within the technical scope protected.
Further, the utilization of resources data of above-mentioned each virtual cpu are specifically as follows the utilization rate (virtual cpu of virtual cpu
Utilization rate be much larger than the utilization rate of load and I/O, both hereafter because can be ignored when actually calculating).
Further, above-mentioned loadtype can include: physical machine loadtype, application cluster loadtype and business defect
At least one of loadtype.
Wherein, physical machine loadtype, application cluster loadtype and business defect loadtype be followed successively by with physical machine,
The loading condition of Service Source is described using, these three levels of virtual cpu.
On this basis, from physical machine loadtype, three angles of application cluster loadtype and business defect loadtype
Degree sets out, can the content further to above-mentioned steps S320 and S330 refine.
Further, in a kind of specific implementation of step S320, can by the utilization of resources data of each virtual cpu with
Physical machine is that unit is counted, to obtain the resource utilization of the virtual cpu in each physical machine, and by the number of results of statistics
According to as the loading statistics under physical machine loadtype.For example, the utilization rate of each virtual cpu is specific in each physical machine
It is how many.
Correspondingly, load balance process condition under physical machine loadtype can be with are as follows: the benefit of each virtual cpu in physical machine
First threshold is all larger than with rate.The first threshold is empirical value, such as 35%.In fact, when all virtual cpus in a certain physical machine
Utilization rate when being both greater than 35%, illustrate the physical machine be assigned on the whole itself resource it is just insufficient, need whole to object
The Service Source of reason machine carries out dilatation.
Correspondingly, in a kind of specific implementation of step S330, it can be for the utilization rate of each virtual cpu in physical machine
It is all larger than the physical machine of first threshold, carries out whole resource capacity expansion and/or resource isolation, and/or will occupy in the physical machine
On virtual cpu in the instance migration of partial virtual CPU to other physical machines.
Have to the mode of physical machine realization dilatation following several: dilatation can be carried out to the resource in entire physical machine, from
And increase the resource of the occupied virtual cpu of running example in the physical machine;It can also be to the example institute run in the physical machine
The resource of occupancy carries out resource isolation, to not be further added by the load capacity of the physical machine;It can also will occupy in the physical machine
On virtual cpu in the instance migration of partial virtual CPU to other physical machines, has example to void to increase in the physical machine
The usage amount of the resource of quasi- CPU.
Further, in the instance migration to other physical machines for the partial virtual CPU that will be occupied in the physical machine
Resource allocation scheme on virtual cpu can also further count in cloud unit, and the utilization rate of virtual cpu is the smallest virtual
CPU, and will be on the smallest virtual cpu of utilization rate of the example and virtual cpu of the partial virtual CPU in the above-mentioned occupancy physical machine
The example of occupancy carries out location swap, to realize local, point-to-point resource is transferred to other use, to guarantee cloud unit on the whole
Stability is consistent, and local migration is also change after all, and migration repeatedly is also unreasonable.
It further, can be by the utilization of resources data of each virtual cpu in another specific implementation of step S320
It is counted as unit of the application of the object of service, to obtain the included example of each application in its occupied virtual cpu
Resource utilization, and using the result data of statistics as the loading statistics under the application cluster loadtype.For example,
The included example of each application, is how many in the utilization rate of the virtual cpu accordingly occupied.
Correspondingly, load balance process condition under application cluster loadtype can be with are as follows:
Using the example for being included in the utilization rate of its virtual cpu occupied, more than the example of fixed proportion, occupy
The utilization rate of virtual cpu be all larger than second threshold.For example, in an included example of application, the example more than 10%,
The utilization rate for the virtual cpu that should be occupied has been more than second threshold, and such as 45%.In fact, the example included when a certain application
In, there is the example of the ratio greater than 10%, when the utilization rate of the virtual cpu occupied is both greater than 45%, illustrates this using whole
The resource that itself is assigned on body is just insufficient, and the whole Service Source to the application is needed to carry out dilatation.
Correspondingly, can be directed to and included example is applied to account at it in another specific implementation of step S330
In the utilization rate of virtual cpu, more than the example of fixed proportion, the utilization rate of the virtual cpu occupied is all larger than the second threshold
This is carried out whole resource capacity expansion using occupied resource by the application of value.
It further, can be by the utilization of resources data of each virtual cpu in another specific implementation of step S320
It is counted as unit of the application of the object of service, to obtain the included example of each application in its occupied virtual cpu
Resource utilization, and using the result data of statistics as the loading statistics under business defect loadtype.For example, each
It is how many in the utilization rate of the virtual cpu accordingly occupied using the example for being included.
Correspondingly, load balance process condition under business defect loadtype can be with are as follows:
Using the example for being included in the utilization rate of its virtual cpu occupied, the occupied virtual cpu of same instance
Utilization rate is persistently greater than third threshold value.For example, there are fixed certain node instances in an included example of application,
The utilization rate of the virtual cpu of occupancy can not have always been high any more, and be continued above third threshold value, and such as 80%.In fact, when a certain
Using some example for being included, the utilization rate of the virtual cpu occupied is in the state greater than 80% for a long time, illustrates the reality
There may be business defects for example itself, such as there is defect from business code layer face, thus lead to its practical virtual cpu occupied
Utilization rate utilization rate is much bigger than expected.
Correspondingly, can exist for the included example of the application in another specific implementation of step S330
In the utilization rate of its virtual cpu occupied, the utilization rate of the occupied virtual cpu of same instance is persistently greater than third threshold value
Application, there are business defects for the example for determining in the application, and close the corresponding service of the example, to carry out defect row
It looks into.
It when determining that business defect occurs in example itself, needs to close the corresponding service of the example, carries out defect investigation, it is excellent
Change example, fundamentally to solve the problems, such as high load.
The load-balancing method of limited resources provided in an embodiment of the present invention, by each in the physical machine on cloud unit
The utilization of resources data of virtual cpu carry out data statistics by different loadtypes, to obtain corresponding under different loads type
Loading statistics, then to loading statistics carry out respective load type under load balance process condition criterion, such as
Fruit meets load balance process condition, then be adapted with the loadtype to the corresponding Service Source of the loading statistics
Resource allocation, to realize the load balancing of limited cloud service resource.
Embodiment two
Based on the scheme thought of the above-mentioned load balancing for carrying out limited resources by loadtype, as shown in Figure 3b, for this
The load-balancing method flowchart 2 of limited resources shown in inventive embodiments, this method has been done on the basis of Fig. 3 a to be changed a little
It is dynamic.As shown in Figure 3b, the load-balancing method of the limited resources includes the following steps:
S340 obtains the utilization of resources data of at least one virtual cpu on target physical machine;It is empty in each physical machine
It is quasi- to dissolve multiple virtual cpus, for providing Service Source to specified application cluster.
Essentially identical with the implementation procedure of step S310 in this step, distinctive points are, the limited resources in the present embodiment
It is not limited to cloud unit, further includes the physical machine on non-cloud unit, i.e., the resource of the virtual cpu on all target physical machines all may be used
Using the process object as load balancing, and the resource that can choose any number of virtual cpu carries out the place of load balancing
Reason.
S350 carries out data statistics by loadtype to the utilization of resources data of at least one above-mentioned virtual cpu, to obtain
Corresponding loading statistics under different loads type;
The implementation procedure of this step hair can be found in the content of step S320.
S360 carries out and the loadtype the corresponding Service Source of the loading statistics under same loadtype
Adaptable resource allocation.
This step eliminates the judgement about load balancing condition when determining resource allocation compared with step S330
Journey, i.e., Service Source can be carried out under any circumstance corresponding utilization of resources data loadtype be adapted
Resource allocation, specific allocation process can be found in the content of step S320.
In addition, being applied equally in the method for the present embodiment, herein not about method and steps other in embodiment one
It repeats.
The load-balancing method of limited resources provided in an embodiment of the present invention, by least one on target physical machine
The utilization of resources data of virtual cpu carry out data statistics by different loadtypes, to obtain corresponding under different loads type
Loading statistics, then the corresponding Service Source of loading statistics under same loadtype is carried out negative with this
The adaptable resource allocation of type is carried, to realize the load balancing of limited resources.
Embodiment three
It as shown in fig. 4 a, is the load balancing apparatus structure chart one of the limited resources of the embodiment of the present invention, the limited resources
Load balancing apparatus can be used for executing method and step as shown in Figure 3a comprising:
First resource data acquisition module 410, each virtual cpu in the physical machine of same cloud unit belonging to obtaining
Utilization of resources data;The cloud unit by multiple physics units at, be virtualized out multiple virtual cpus in each physical machine, one
A cloud unit is used to provide Service Source to specified application cluster;
First load data statistical module 420 presses preset loadtype for the utilization of resources data to each virtual cpu
Data statistics is carried out, to obtain corresponding loading statistics under different loads type;
First resource scheduling module 430, if meeting the load balancing under respective load type for loading statistics
Treatment conditions then carry out the resource allocation being adapted with the loadtype to the corresponding Service Source of the loading statistics.
Further, the utilization of resources data of above-mentioned each virtual cpu are the utilization rate of virtual cpu.
Further, above-mentioned loadtype includes: that physical machine loadtype, application cluster loadtype and business defect are negative
Carry at least one of type.
On this basis, the first load data statistical module 420 can include:
Physical machine load data statistic unit, for by the utilization of resources data of each virtual cpu as unit of physical machine into
Row statistics, to obtain the resource utilization of the virtual cpu in each physical machine, and the result data of statistics is born as physical machine
Carry the loading statistics under type.
Correspondingly, load balance process condition under the physical machine loadtype can be with are as follows: each virtual cpu in physical machine
Utilization rate be all larger than first threshold.
Correspondingly, first resource scheduling module 430 can include:
Physical machine resource allocation unit, for being all larger than first threshold for the utilization rate of each virtual cpu in physical machine
Physical machine, carries out whole resource capacity expansion and/or resource isolation, and/or by the example of the partial virtual CPU occupied in the physical machine
It moves on the virtual cpu in other physical machines.
Further, the physical machine resource allocation unit is specifically used for, and counts in cloud unit, the utilization rate of virtual cpu
The smallest virtual cpu, and by the smallest void of utilization rate of the example and virtual cpu of the partial virtual CPU occupied in the physical machine
The example occupied on quasi- CPU carries out location swap.
Further, the first load data statistical module 420 may also include that
Application cluster load data statistic unit, the object for servicing the utilization of resources data of each virtual cpu are answered
It is counted as unit of, to obtain the included example of each application in the resource utilization of its occupied virtual cpu, and
Using the result data of statistics as the loading statistics under application cluster loadtype.
Correspondingly, load balance process condition under the application cluster loadtype can be with are as follows: apply included reality
Example is in the utilization rate of its virtual cpu occupied, and more than the example of fixed proportion, the utilization rate of the virtual cpu occupied is big
In second threshold.
Correspondingly, first resource scheduling module 430 can include:
Application cluster resource allocation unit, for being directed to the included example of application in the utilization of its virtual cpu occupied
In rate, more than the example of fixed proportion, the utilization rate of the virtual cpu occupied is all larger than the application of second threshold, by the application
Occupied resource carries out whole resource capacity expansion.
Further, the first load data statistical module 420 may also include that
Business defect load data statistic unit, the object for servicing the utilization of resources data of each virtual cpu are answered
It is counted as unit of, to obtain the included example of each application in the resource utilization of its occupied virtual cpu, and
Using the result data of statistics as the loading statistics under the business defect loadtype.
Correspondingly, load balance process condition under the business defect loadtype can be with are as follows: apply included reality
For example in the utilization rate of its virtual cpu occupied, the utilization rate of the occupied virtual cpu of same instance is persistently greater than third threshold
Value.
Correspondingly, first resource scheduling module 430 can include:
Business defect resource allocation unit, for being directed to the included example of application in the utilization of its virtual cpu occupied
In rate, the utilization rate of the occupied virtual cpu of same instance is persistently greater than the application of third threshold value, determines the reality in the application
There are business defects for example, and close the corresponding service of the example, to carry out defect investigation.
The load balancing apparatus of limited resources provided in an embodiment of the present invention, by each in the physical machine on cloud unit
The utilization of resources data of virtual cpu carry out data statistics by different loadtypes, to obtain corresponding under different loads type
Loading statistics, then to loading statistics carry out respective load type under load balance process condition criterion, such as
Fruit meets load balance process condition, then be adapted with the loadtype to the corresponding Service Source of the loading statistics
Resource allocation, to realize the load balancing of limited cloud service resource.
Example IV
It as shown in Figure 4 b, is the load balancing apparatus structure chart two of the limited resources of the embodiment of the present invention, the limited resources
Load balancing apparatus can be used for executing method and step as shown in Figure 3b comprising:
Secondary resource data acquisition module 440, for obtaining the resource benefit of at least one virtual cpu on target physical machine
Use data;Multiple virtual cpus are virtualized out in each physical machine, for providing Service Source to specified application cluster;
Second load data statistical module 450 presses loadtype for the utilization of resources data at least one virtual cpu
Data statistics is carried out, to obtain corresponding loading statistics under different loads type;
Secondary resource scheduling module 460, for the corresponding Service Source of loading statistics under same loadtype
Carry out the resource allocation being adapted with the loadtype.
The load balancing apparatus of limited resources provided in an embodiment of the present invention, by least one on target physical machine
The utilization of resources data of virtual cpu carry out data statistics by different loadtypes, to obtain corresponding under different loads type
Loading statistics, then the corresponding Service Source of loading statistics under same loadtype is carried out negative with this
The adaptable resource allocation of type is carried, to realize the load balancing of limited resources.
Embodiment five
The overall architecture of the load balancing apparatus of limited resources is described in preceding embodiment three, the function of the device can borrow
It helps a kind of electronic equipment to realize to complete, as shown in figure 5, its structural schematic diagram for the electronic equipment of the embodiment of the present invention, specifically
It include: memory 510 and processor 520.
Memory 510, for storing program.
In addition to above procedure, memory 510 is also configured to store various other data to support in electronic equipment
On operation.The example of these data includes the instruction of any application or method for operating on an electronic device, connection
It is personal data, telephone book data, message, picture, video etc..
Memory 510 can realize by any kind of volatibility or non-volatile memory device or their combination,
Such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable is read-only
Memory (EPROM), programmable read only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk
Or CD.
Processor 520 is coupled to memory 510, for executing the program in memory 510, to be used for:
The utilization of resources data of each virtual cpu in the physical machine of same cloud unit belonging to obtaining;The cloud unit is by more
A physics unit is at being virtualized out multiple virtual cpus, a cloud unit is used for specified in each physical machine
Application cluster provides Service Source;
Data statistics is carried out by preset loadtype to the utilization of resources data of each virtual cpu, to obtain difference
Corresponding loading statistics under loadtype;
If the loading statistics meet the load balance process condition under respective load type, unite to the load
It counts corresponding Service Source and carries out the resource allocation being adapted with the loadtype.
Above-mentioned specific processing operation is described in detail in embodiment in front, and details are not described herein.
Further, as shown in figure 5, electronic equipment can also include: communication component 530, power supply module 540, audio component
550, other components such as display 560.Members are only schematically provided in Fig. 5, are not meant to that electronic equipment only includes Fig. 5
Shown component.
Communication component 530 is configured to facilitate the communication of wired or wireless way between electronic equipment and other equipment.Electricity
Sub- equipment can access the wireless network based on communication standard, such as WiFi, 2G or 3G or their combination.It is exemplary at one
In embodiment, communication component 530 receives broadcast singal or broadcast correlation from external broadcasting management system via broadcast channel
Information.In one exemplary embodiment, communication component 530 further includes near-field communication (NFC) module, to promote short range communication.
For example, radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra wide band (UWB) skill can be based in NFC module
Art, bluetooth (BT) technology and other technologies are realized.
Power supply module 540 provides electric power for the various assemblies of electronic equipment.Power supply module 540 may include power management
System, one or more power supplys and other with for electronic equipment generate, manage, and distribute the associated component of electric power.
Audio component 550 is configured as output and/or input audio signal.For example, audio component 550 includes a Mike
Wind (MIC), when electronic equipment is in operation mode, when such as call mode, recording mode, and voice recognition mode, microphone is matched
It is set to reception external audio signal.The received audio signal can be further stored in memory 510 or via communication set
Part 530 is sent.In some embodiments, audio component 550 further includes a loudspeaker, is used for output audio signal.
Display 560 includes screen, and screen may include liquid crystal display (LCD) and touch panel (TP).If screen
Curtain includes touch panel, and screen may be implemented as touch screen, to receive input signal from the user.Touch panel includes one
A or multiple touch sensors are to sense the gesture on touch, slide, and touch panel.Touch sensor can not only sense touching
It touches or the boundary of sliding action, but also detects duration and pressure relevant with touch or slide.
Embodiment six
The overall architecture of the load balancing apparatus of limited resources is described in preceding embodiment four, the function of the device can borrow
It helps a kind of electronic equipment to realize to complete, as shown in fig. 6, its structural schematic diagram for the electronic equipment of the embodiment of the present invention, specifically
It include: memory 610 and processor 620.
Memory 610, for storing program.
In addition to above procedure, memory 610 is also configured to store various other data to support in electronic equipment
On operation.The example of these data includes the instruction of any application or method for operating on an electronic device, connection
It is personal data, telephone book data, message, picture, video etc..
Memory 610 can realize by any kind of volatibility or non-volatile memory device or their combination,
Such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable is read-only
Memory (EPROM), programmable read only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk
Or CD.
Processor 620 is coupled to memory 610, for executing the program in memory 610, to be used for:
Obtain the utilization of resources data of at least one virtual cpu on target physical machine;It is empty in each physical machine
It is quasi- to dissolve multiple virtual cpus, for providing Service Source to specified application cluster;
Data statistics is carried out by loadtype to the utilization of resources data of at least one virtual cpu, to obtain difference
Corresponding loading statistics under loadtype;
The corresponding Service Source of the loading statistics under same loadtype mutually fit with the loadtype
The resource allocation answered.
Above-mentioned specific processing operation is described in detail in embodiment in front, and details are not described herein.
Further, as shown in fig. 6, electronic equipment can also include: communication component 630, power supply module 640, audio component
650, other components such as display 660.Members are only schematically provided in Fig. 6, are not meant to that electronic equipment only includes Fig. 6
Shown component.
Communication component 630 is configured to facilitate the communication of wired or wireless way between electronic equipment and other equipment.Electricity
Sub- equipment can access the wireless network based on communication standard, such as WiFi, 2G or 3G or their combination.It is exemplary at one
In embodiment, communication component 630 receives broadcast singal or broadcast correlation from external broadcasting management system via broadcast channel
Information.In one exemplary embodiment, communication component 630 further includes near-field communication (NFC) module, to promote short range communication.
For example, radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra wide band (UWB) skill can be based in NFC module
Art, bluetooth (BT) technology and other technologies are realized.
Power supply module 640 provides electric power for the various assemblies of electronic equipment.Power supply module 640 may include power management
System, one or more power supplys and other with for electronic equipment generate, manage, and distribute the associated component of electric power.
Audio component 650 is configured as output and/or input audio signal.For example, audio component 650 includes a Mike
Wind (MIC), when electronic equipment is in operation mode, when such as call mode, recording mode, and voice recognition mode, microphone is matched
It is set to reception external audio signal.The received audio signal can be further stored in memory 610 or via communication set
Part 630 is sent.In some embodiments, audio component 650 further includes a loudspeaker, is used for output audio signal.
Display 660 includes screen, and screen may include liquid crystal display (LCD) and touch panel (TP).If screen
Curtain includes touch panel, and screen may be implemented as touch screen, to receive input signal from the user.Touch panel includes one
A or multiple touch sensors are to sense the gesture on touch, slide, and touch panel.Touch sensor can not only sense touching
It touches or the boundary of sliding action, but also detects duration and pressure relevant with touch or slide.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above-mentioned each method embodiment can lead to
The relevant hardware of program instruction is crossed to complete.Program above-mentioned can be stored in a computer readable storage medium.The journey
When being executed, execution includes the steps that above-mentioned each method embodiment to sequence;And storage medium above-mentioned include: ROM, RAM, magnetic disk or
The various media that can store program code such as person's CD.
Finally, it should be noted that the above various embodiments is only to illustrate the technical solution of the application, rather than its limitations;To the greatest extent
Pipe is described in detail the application referring to foregoing embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement;And these are modified or replaceed, each embodiment technology of the application that it does not separate the essence of the corresponding technical solution
The range of scheme.
Claims (15)
1. a kind of load-balancing method of limited resources characterized by comprising
The utilization of resources data of each virtual cpu in the physical machine of same cloud unit belonging to obtaining;The cloud unit is by multiple objects
Reason machine forms, and multiple virtual cpus are virtualized out in each physical machine, and a cloud unit is used for specified application
Cluster provides Service Source;
Data statistics is carried out by preset loadtype to the utilization of resources data of each virtual cpu, to obtain different loads
Corresponding loading statistics under type;
If the loading statistics meet the load balance process condition under respective load type, to the load statistics number
The resource allocation being adapted with the loadtype is carried out according to corresponding Service Source.
2. the method according to claim 1, wherein the utilization of resources data of each virtual cpu are virtual cpu
Utilization rate.
3. according to the method described in claim 2, it is characterized in that, the loadtype includes: physical machine loadtype, application
At least one of cluster loadtype and business defect loadtype.
4. according to the method described in claim 3, it is characterized in that, the utilization of resources data to each virtual cpu are pressed
Preset loadtype carries out data statistics, includes: to obtain corresponding loading statistics under different loads type
The utilization of resources data of each virtual cpu are counted as unit of physical machine, to obtain in each physical machine
Virtual cpu resource utilization, and using the result data of statistics as the load statistics under the physical machine loadtype
Data.
5. according to the method described in claim 4, it is characterized in that, load balance process item under the physical machine loadtype
Part are as follows:
The utilization rate of each virtual cpu is all larger than first threshold in the physical machine;
The resource allocation that the corresponding Service Source of loading statistics is carried out to be adapted with the loadtype, comprising:
It is all larger than the physical machine of first threshold for the utilization rate of each virtual cpu in the physical machine, carries out whole resource capacity expansion
And/or resource isolation, and/or by the void in the instance migration to other physical machines of the partial virtual CPU occupied in the physical machine
On quasi- CPU.
6. according to the method described in claim 5, it is characterized in that, described by the partial virtual CPU's occupied in the physical machine
Include: on virtual cpu in instance migration to other physical machines
Count in the cloud unit, the smallest virtual cpu of the utilization rate of virtual cpu, and by it is described occupancy the physical machine on portion
The example occupied on the example of virtual cpu and the smallest virtual cpu of utilization rate of the virtual cpu is divided to carry out location swap.
7. according to the method described in claim 3, it is characterized in that, the utilization of resources data to each virtual cpu are pressed
Preset loadtype carries out data statistics, includes: to obtain corresponding loading statistics under different loads type
The utilization of resources data of each virtual cpu are counted as unit of the object application serviced, to obtain each application
The example for being included its occupied virtual cpu resource utilization, and using the result data of statistics as described in
Loading statistics under application cluster loadtype.
8. the method according to the description of claim 7 is characterized in that the load balance process under the application cluster loadtype
Condition are as follows:
The included example of the application is in the utilization rate of its virtual cpu occupied, more than the example of fixed proportion,
The utilization rate of the virtual cpu of occupancy is all larger than second threshold.
The resource allocation that the corresponding Service Source of loading statistics is carried out to be adapted with the loadtype, comprising:
For the included example of the application in the utilization rate of its virtual cpu occupied, more than the reality of fixed proportion
The utilization rate of example, the virtual cpu occupied is all larger than the application of second threshold, this is carried out whole money using occupied resource
Source dilatation.
9. according to the method described in claim 3, it is characterized in that, the utilization of resources data to each virtual cpu are pressed
Preset loadtype carries out data statistics, includes: to obtain corresponding loading statistics under different loads type
The utilization of resources data of each virtual cpu are counted as unit of the object application serviced, to obtain each application
The example for being included its occupied virtual cpu resource utilization, and using the result data of statistics as described in
Loading statistics under business defect loadtype.
10. according to the method described in claim 9, it is characterized in that, at load balancing under the business defect loadtype
Manage bar part are as follows:
For the included example of the application in the utilization rate of its virtual cpu occupied, same instance is occupied virtual
The utilization rate of CPU is persistently greater than third threshold value;
The resource allocation that the corresponding Service Source of loading statistics is carried out to be adapted with the loadtype, comprising:
For the included example of the application in the utilization rate of its virtual cpu occupied, same instance is occupied
The utilization rate of virtual cpu is persistently greater than the application of third threshold value, and there are business defects for the example for determining in the application, and close
The corresponding service of the example, to carry out defect investigation.
11. a kind of load-balancing method of limited resources characterized by comprising
Obtain the utilization of resources data of at least one virtual cpu on target physical machine;It is virtualized in each physical machine
Multiple virtual cpus out, for providing Service Source to specified application cluster;
Data statistics is carried out by loadtype to the utilization of resources data of at least one virtual cpu, to obtain different loads
Corresponding loading statistics under type;
The corresponding Service Source progress of the loading statistics under same loadtype is adapted with the loadtype
Resource allocation.
12. a kind of load balancing apparatus of limited resources characterized by comprising
First resource data acquisition module, the resource benefit of each virtual cpu in physical machine for obtaining affiliated same cloud unit
Use data;The cloud unit by multiple physics units at, be virtualized out multiple virtual cpus in each physical machine, one
The cloud unit is used to provide Service Source to specified application cluster;
First load data statistical module, for the utilization of resources data to each virtual cpu by preset loadtype into
Line number according to statistics, to obtain corresponding loading statistics under different loads type;
First resource scheduling module, if meeting the load balance process under respective load type for the loading statistics
Condition then carries out the resource allocation being adapted with the loadtype to the corresponding Service Source of the loading statistics.
13. a kind of load balancing apparatus of limited resources characterized by comprising
Secondary resource data acquisition module, for obtaining the utilization of resources data of at least one virtual cpu on target physical machine;
Multiple virtual cpus are virtualized out in each physical machine, for providing Service Source to specified application cluster;
Second load data statistical module, for the utilization of resources data at least one virtual cpu by loadtype into
Line number according to statistics, to obtain corresponding loading statistics under different loads type;
Secondary resource scheduling module, for being carried out to the corresponding Service Source of the loading statistics under same loadtype
The resource allocation being adapted with the loadtype.
14. a kind of electronic equipment characterized by comprising
Memory, for storing program;
Processor is coupled to the memory, for executing described program, to be used for:
The utilization of resources data of each virtual cpu in the physical machine of same cloud unit belonging to obtaining;The cloud unit is by multiple objects
Reason machine forms, and multiple virtual cpus are virtualized out in each physical machine, and a cloud unit is used for specified application
Cluster provides Service Source;
Data statistics is carried out by preset loadtype to the utilization of resources data of each virtual cpu, to obtain different loads
Corresponding loading statistics under type;
If the loading statistics meet the load balance process condition under respective load type, to the load statistics number
The resource allocation being adapted with the loadtype is carried out according to corresponding Service Source.
15. a kind of electronic equipment characterized by comprising
Memory, for storing program;
Processor is coupled to the memory, for executing described program, to be used for:
Obtain the utilization of resources data of at least one virtual cpu on target physical machine;It is virtualized in each physical machine
Multiple virtual cpus out, for providing Service Source to specified application cluster;
Data statistics is carried out by loadtype to the utilization of resources data of at least one virtual cpu, to obtain different loads
Corresponding loading statistics under type;
The corresponding Service Source progress of the loading statistics under same loadtype is adapted with the loadtype
Resource allocation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710499500.8A CN109144658B (en) | 2017-06-27 | 2017-06-27 | Load balancing method and device for limited resources and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710499500.8A CN109144658B (en) | 2017-06-27 | 2017-06-27 | Load balancing method and device for limited resources and electronic equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109144658A true CN109144658A (en) | 2019-01-04 |
CN109144658B CN109144658B (en) | 2022-07-15 |
Family
ID=64805043
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710499500.8A Active CN109144658B (en) | 2017-06-27 | 2017-06-27 | Load balancing method and device for limited resources and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109144658B (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110545258A (en) * | 2019-07-25 | 2019-12-06 | 浙江大华技术股份有限公司 | Streaming media server resource allocation method and device and server |
CN110673928A (en) * | 2019-09-29 | 2020-01-10 | 天津卓朗科技发展有限公司 | Thread binding method, thread binding device, storage medium and server |
CN111400049A (en) * | 2020-03-26 | 2020-07-10 | 北京搜房科技发展有限公司 | Resource adjusting method and device |
CN111427673A (en) * | 2020-03-16 | 2020-07-17 | 杭州迪普科技股份有限公司 | Load balancing method, device and equipment |
WO2021139264A1 (en) * | 2020-07-28 | 2021-07-15 | 平安科技(深圳)有限公司 | Object storage control method and apparatus, computer device and storage medium |
WO2022247189A1 (en) * | 2021-05-24 | 2022-12-01 | 北京灵汐科技有限公司 | Core control method and apparatus for many-core system, and many-core system |
CN115529242A (en) * | 2022-09-23 | 2022-12-27 | 浙江大学 | Method for realizing cloud network resource allocation under optimal water level |
CN117687799A (en) * | 2024-02-02 | 2024-03-12 | 中国科学院空天信息创新研究院 | Distributed stream type acceleration method and computing terminal for remote sensing interpretation application |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101504620A (en) * | 2009-03-03 | 2009-08-12 | 华为技术有限公司 | Load balancing method, apparatus and system of virtual cluster system |
CN102096461A (en) * | 2011-01-13 | 2011-06-15 | 浙江大学 | Energy-saving method of cloud data center based on virtual machine migration and load perception integration |
CN102710503A (en) * | 2012-05-15 | 2012-10-03 | 浪潮电子信息产业股份有限公司 | Network load balancing method based on cloud sea OS (operation system) |
CN103955398A (en) * | 2014-04-28 | 2014-07-30 | 浙江大学 | Virtual machine coexisting scheduling method based on processor performance monitoring |
CN104102543A (en) * | 2014-06-27 | 2014-10-15 | 北京奇艺世纪科技有限公司 | Load regulation method and load regulation device in cloud computing environment |
CN104166594A (en) * | 2014-08-19 | 2014-11-26 | 杭州华为数字技术有限公司 | Load balancing control method and related devices |
CN104375897A (en) * | 2014-10-27 | 2015-02-25 | 西安工程大学 | Cloud computing resource scheduling method based on minimum relative load imbalance degree |
CN104484220A (en) * | 2014-11-28 | 2015-04-01 | 杭州华为数字技术有限公司 | Method and device for dispatching dynamic resources of virtual cluster |
CN104836819A (en) * | 2014-02-10 | 2015-08-12 | 阿里巴巴集团控股有限公司 | Dynamic load balancing method and system, and monitoring and dispatching device |
CN104850461A (en) * | 2015-05-12 | 2015-08-19 | 华中科技大学 | NUMA-oriented virtual cpu (central processing unit) scheduling and optimizing method |
US20160085571A1 (en) * | 2014-09-21 | 2016-03-24 | Vmware, Inc. | Adaptive CPU NUMA Scheduling |
-
2017
- 2017-06-27 CN CN201710499500.8A patent/CN109144658B/en active Active
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101504620A (en) * | 2009-03-03 | 2009-08-12 | 华为技术有限公司 | Load balancing method, apparatus and system of virtual cluster system |
CN102096461A (en) * | 2011-01-13 | 2011-06-15 | 浙江大学 | Energy-saving method of cloud data center based on virtual machine migration and load perception integration |
CN102710503A (en) * | 2012-05-15 | 2012-10-03 | 浪潮电子信息产业股份有限公司 | Network load balancing method based on cloud sea OS (operation system) |
CN104836819A (en) * | 2014-02-10 | 2015-08-12 | 阿里巴巴集团控股有限公司 | Dynamic load balancing method and system, and monitoring and dispatching device |
CN103955398A (en) * | 2014-04-28 | 2014-07-30 | 浙江大学 | Virtual machine coexisting scheduling method based on processor performance monitoring |
CN104102543A (en) * | 2014-06-27 | 2014-10-15 | 北京奇艺世纪科技有限公司 | Load regulation method and load regulation device in cloud computing environment |
CN104166594A (en) * | 2014-08-19 | 2014-11-26 | 杭州华为数字技术有限公司 | Load balancing control method and related devices |
US20160085571A1 (en) * | 2014-09-21 | 2016-03-24 | Vmware, Inc. | Adaptive CPU NUMA Scheduling |
CN104375897A (en) * | 2014-10-27 | 2015-02-25 | 西安工程大学 | Cloud computing resource scheduling method based on minimum relative load imbalance degree |
CN104484220A (en) * | 2014-11-28 | 2015-04-01 | 杭州华为数字技术有限公司 | Method and device for dispatching dynamic resources of virtual cluster |
CN104850461A (en) * | 2015-05-12 | 2015-08-19 | 华中科技大学 | NUMA-oriented virtual cpu (central processing unit) scheduling and optimizing method |
Non-Patent Citations (2)
Title |
---|
YUXIA CHENG等: "A User-Level NUMA-Aware Scheduler for Optimizing Virtual Machine Performance", 《10TH INTERNATIONAL SYMPOSIUM,APPT2013》 * |
余超 等: "A-可负载均衡的实时虚拟机VCPU调度算法", 《华中科技大学学报(自然科学版)》 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110545258A (en) * | 2019-07-25 | 2019-12-06 | 浙江大华技术股份有限公司 | Streaming media server resource allocation method and device and server |
CN110673928A (en) * | 2019-09-29 | 2020-01-10 | 天津卓朗科技发展有限公司 | Thread binding method, thread binding device, storage medium and server |
CN110673928B (en) * | 2019-09-29 | 2021-12-14 | 天津卓朗科技发展有限公司 | Thread binding method, thread binding device, storage medium and server |
CN111427673A (en) * | 2020-03-16 | 2020-07-17 | 杭州迪普科技股份有限公司 | Load balancing method, device and equipment |
CN111427673B (en) * | 2020-03-16 | 2023-04-07 | 杭州迪普科技股份有限公司 | Load balancing method, device and equipment |
CN111400049A (en) * | 2020-03-26 | 2020-07-10 | 北京搜房科技发展有限公司 | Resource adjusting method and device |
WO2021139264A1 (en) * | 2020-07-28 | 2021-07-15 | 平安科技(深圳)有限公司 | Object storage control method and apparatus, computer device and storage medium |
WO2022247189A1 (en) * | 2021-05-24 | 2022-12-01 | 北京灵汐科技有限公司 | Core control method and apparatus for many-core system, and many-core system |
CN115529242A (en) * | 2022-09-23 | 2022-12-27 | 浙江大学 | Method for realizing cloud network resource allocation under optimal water level |
CN115529242B (en) * | 2022-09-23 | 2023-07-18 | 浙江大学 | Method for realizing cloud network resource allocation under optimal water level |
CN117687799A (en) * | 2024-02-02 | 2024-03-12 | 中国科学院空天信息创新研究院 | Distributed stream type acceleration method and computing terminal for remote sensing interpretation application |
Also Published As
Publication number | Publication date |
---|---|
CN109144658B (en) | 2022-07-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109144658A (en) | Load-balancing method, device and the electronic equipment of limited resources | |
CN106104485A (en) | Dynamic resource management for multi-process application | |
WO2017133563A1 (en) | Method and apparatus for establishing associative relationship | |
US10009213B2 (en) | System and method for isolation of multi-tenant platform customization using child processes | |
CN104620222A (en) | Scaling a virtual machine instance | |
CN102866903A (en) | Decoupling backstage work with forestage work | |
CN106662909A (en) | Heuristic processsor power management in operating systems | |
US10614676B2 (en) | Monitoring cash supply-related information and managing refill of a cash supply | |
CN109951545A (en) | The emerging system of adaptive container and cloud desktop and its method for obtaining cloud resource | |
CN107133087A (en) | A kind of resource regulating method and equipment | |
Grosskop et al. | Identification of application-level energy optimizations | |
CN106302628B (en) | Unified management scheduling method for computing resources in ARM architecture network cluster | |
Han et al. | Refining microservices placement employing workload profiling over multiple kubernetes clusters | |
CN109840139A (en) | Method, apparatus, electronic equipment and the storage medium of resource management | |
WO2023174037A1 (en) | Resource scheduling method, apparatus and system, device, medium, and program product | |
CN115345464A (en) | Service order dispatching method and device, computer equipment and storage medium | |
CN109657893A (en) | Business datum distribution method, device, equipment and computer readable storage medium | |
CN109582439A (en) | DCN dispositions method, device, equipment and computer readable storage medium | |
CN110032750A (en) | A kind of model construction, data life period prediction technique, device and equipment | |
CN109753353A (en) | Resources of virtual machine distribution method, device and electronic equipment | |
US11544589B2 (en) | Use machine learning to verify and modify a specification of an integration interface for a software module | |
CN109960572B (en) | Equipment resource management method and device and intelligent terminal | |
US20220413933A1 (en) | Liaison System and Method for Cloud Computing Environment | |
CN107634978A (en) | A kind of resource regulating method and device | |
US20220147380A1 (en) | Optimizing Hybrid Cloud Usage |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20230526 Address after: Room 1-2-A06, Yungu Park, No. 1008 Dengcai Street, Sandun Town, Xihu District, Hangzhou City, Zhejiang Province, 310030 Patentee after: Aliyun Computing Co.,Ltd. Address before: Box 847, four, Grand Cayman capital, Cayman Islands, UK Patentee before: ALIBABA GROUP HOLDING Ltd. |