CN106713375A - Method and device for allocating cloud resources - Google Patents
Method and device for allocating cloud resources Download PDFInfo
- Publication number
- CN106713375A CN106713375A CN201510430509.4A CN201510430509A CN106713375A CN 106713375 A CN106713375 A CN 106713375A CN 201510430509 A CN201510430509 A CN 201510430509A CN 106713375 A CN106713375 A CN 106713375A
- Authority
- CN
- China
- Prior art keywords
- cloud
- resource
- domain
- cloud domain
- resource requirement
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a method and a device for allocating cloud resources. The method is applied to a cloud platform comprising a first cloud domain and a second cloud domain. The method comprises steps: a resource demand of an account in the first cloud domain is acquired; based on the resource demand, whether the first cloud domain can provide enough resources is judged; and if the first cloud domain can not provide enough resources, at least part of idle resources in the second cloud domain are reallocated to the first cloud domain based on the resource demand, wherein the first cloud domain is a private cloud domain and the second cloud domain is a public cloud domain, or the first cloud domain is a public cloud domain and the second cloud domain is a private cloud domain.
Description
Technical field
The present invention relates to field of cloud computer technology, more particularly to a kind of cloud Resource allocation and smoothing method and device.
Background technology
With the development of cloud computing, the use of cloud platform and cloud computing is more and more frequent.Cloud resource can typically divide
It is public cloud resource and private clound resource.
Public cloud refers to by the third party provider standard cloud computing service that account is provided towards the public, user Ke Tong
The infrastructure crossed needed for internet is directly accessed publicly-owned cloud platform acquisition is serviced
(Infrastructure as a Service, IAAS), platform are service (Platform-as-a-Service, PAAS)
And software is service (Software as a Service, SAAS).Public cloud can provide quick friendship for account
The ability pay, consumed on demand.
Private clound is generally deployed in inside fire wall, is individually built in enterprises, is the various letters in unit inside
Breath technological system provides the resource services such as required IAAS, PAAS and SAAS.Private clound is various to integrating
Information technology resources, increasing operation rate has many advantages.
Public cloud and private clound are the cloud computing service presentation mode of comparative maturity, but individually build private clound
With public cloud high expensive, it is impossible to effective cost efficiency.In order to improve the basis of cloud platform overall utilization rate
Upper to realize profit, from the framework of mixed cloud, mixing cloud system is in private clound foundation construction for more companies
On, it is supplied to external user to use the corresponding cloud resource of private clound by Network Isolation.But mixed cloud is such as
How He Gengyou rationally to private clound domain and public cloud domain distribution cloud resource and enters the resource allocation that racks.
The content of the invention
In view of this, the embodiment of the present invention is expected to provide a kind of cloud resource allocation method and device, can be at least
Part solves the problems, such as the cloud resource allocation between public cloud domain and private clound domain in mixed cloud.
To reach above-mentioned purpose, the technical proposal of the invention is realized in this way：
Embodiment of the present invention first aspect provides a kind of cloud Resource allocation and smoothing method, and methods described is applied to include
In the cloud platform in the first cloud domain and the second cloud domain, methods described includes：
Obtain the resource requirement of the account in the first cloud domain；
Based on the resource requirement, judge whether the first cloud domain provides enough resources；
If the first cloud domain can not provide enough resources, based on the resource requirement by second cloud
The slack resources in domain are at least partly reassigned to the first cloud domain；
Wherein, the first cloud domain is private clound domain, and the second cloud domain is public cloud domain；Or, described
One cloud domain is public cloud domain, and the second cloud domain is private clound domain.
Based on such scheme, methods described also includes：
Determine the priority parameters of the resource requirement；
Based on the priority parameters, corresponding account is given by the resource allocation in the first cloud domain, to meet
The resource requirement.
Based on such scheme, the priority parameters of the determination resource requirement, including：
Calculate the application weights of the resource requirement；
Application time based on the resource requirement, the application weights and preset function relation, calculate described
Priority parameters.
Based on such scheme, the application weights of the calculating resource requirement, including：
Based on the resource type parameter needed for the corresponding account level of the resource requirement, the resource requirement and
Resource quantity parameter needed for the resource requirement at least one, calculate the application weights.
Based on such scheme, the resource requirement of the account in the first cloud of acquisition domain, including：
In the first resource requirement for specifying acquisition account in the time；
It is described to judge whether the first cloud domain provides enough resources based on the resource requirement, including：
Specified in the time second, based on the resource requirement of account each described, judge first cloud
Whether domain provides enough resources；
Described second specifies the time to be later than described second specifies the time.
Based on such scheme, if the first cloud domain can not provide enough resources, based on the money
Source demand is at least partly reassigned to the first cloud domain by the slack resources in the second cloud domain, including：
Count the second cloud domain idle PAAS resources and/or SAAS resources；
Based on the resource requirement, the idle PAAS resources and SAAS resources are allocated to described
One cloud domain.
Embodiment of the present invention second aspect provides a kind of cloud Resource allocation and smoothing device, and described device is applied to include
In the cloud platform in the first cloud domain and the second cloud domain, described device includes：
Acquiring unit, the resource requirement of the account for obtaining the first cloud domain；
Judging unit, for based on the resource requirement, judging whether the first cloud domain provides enough money
Source；
Deployment unit, if can not provide enough resources for the first cloud domain, is needed based on the resource
Ask and the slack resources in the second cloud domain are at least partly reassigned to the first cloud domain；
Wherein, the first cloud domain is private clound domain, and the second cloud domain is public cloud domain；Or, described
One cloud domain is public cloud domain, and the second cloud domain is private clound domain.
Based on such scheme, described device also includes：
Determining unit, the priority parameters for determining the resource requirement；
The deployment unit, is additionally operable to based on the priority parameters, by the resource allocation in the first cloud domain
To corresponding account, to meet the resource requirement.
Based on such scheme, the determining unit, the application weights for calculating the resource requirement；It is based on
The application time of the resource requirement, the application weights and preset function relation, calculate the priority ginseng
Number.
Based on such scheme, the determining unit, it is specific with based on the corresponding account level of the resource requirement,
The resource quantity parameter needed for resource type parameter and the resource requirement needed for the resource requirement at least its
One of, calculate the application weights.
Based on such scheme, the acquiring unit, specifically for specifying the money that account is obtained in the time first
Source demand；
The judging unit, specifically for being specified in the time second, the money based on account each described
Source demand, judges whether the first cloud domain provides enough resources；Described second specified time was later than described
Second specifies the time.
Based on such scheme, the deployment unit, the PAAS idle specifically for counting the second cloud domain
Resource and/or SAAS resources；
Based on the resource requirement, the idle PAAS resources and SAAS resources are allocated to described
One cloud domain.
Cloud resource allocation method and device described in the embodiment of the present invention, understand the resource requirement according to the first cloud domain,
Judge the enough resources whether the first cloud domain can provide, if can not if will be would sit idle for from the second cloud domain money
Source is divided into the first cloud domain, is first that a kind of dynamic resource is adjusted like this to meet resource requirement
Method of completing the square, it is clear that meet resource requirement, utilization resource more efficiently；Can be to the first cloud in mixed cloud
The cloud resource in domain and the second cloud domain carries out unified allocation of resources.
Brief description of the drawings
Fig. 1 is one of schematic flow sheet of dispatching method of cloud resource described in the embodiment of the present invention；
Fig. 2 is the two of the schematic flow sheet of the dispatching method of the cloud resource described in the embodiment of the present invention；
Fig. 3 is the structural representation of the dispatching device of the cloud resource described in the embodiment of the present invention；
Fig. 4 is a kind of structural representation of mixed cloud provided in an embodiment of the present invention；
Fig. 5 is the three of the schematic flow sheet of the dispatching method of the cloud resource described in the embodiment of the present invention；
Fig. 6 is the four of the schematic flow sheet of the dispatching method of the cloud resource described in the embodiment of the present invention；
Fig. 7 is the five of the schematic flow sheet of the dispatching method of the cloud resource described in the embodiment of the present invention；
Fig. 8 is a kind of structural representation of database resource provided in an embodiment of the present invention；
Fig. 9 is the schematic flow sheet of database migration provided in an embodiment of the present invention.
Specific embodiment
Technical scheme is done below in conjunction with Figure of description and specific embodiment is further explained in detail
State.
Embodiment of the method：
As shown in figure 1, the present embodiment provides a kind of cloud Resource allocation and smoothing method, methods described is applied to include
In the cloud platform in the first cloud domain and the second cloud domain, methods described includes：
Step S110：Obtain the resource requirement of the account in the first cloud domain；
Step S120：Based on the resource requirement, judge whether the first cloud domain provides enough resources；
Step S130：If the first cloud domain can not provide enough resources, will based on the resource requirement
The slack resources in the second cloud domain are at least partly reassigned to the first cloud domain；
Wherein, the first cloud domain is private clound domain, and the second cloud domain is public cloud domain；Or, described
One cloud domain is public cloud domain, and the second cloud domain is private clound domain.
Cloud platform described in the present embodiment can be in the mixing cloud platform in private clound domain and public cloud domain.The cloud
Platform can provide various fortune resources.
Step S110 described in the present embodiment is obtained and obtained by receiving the capacity expansion application of each account；Also may be used
Obtained with by monitoring the cloud service condition of each account.
The account of public cloud domain described in the present embodiment is external account.The account in the private clound domain is mixing
The content account of cloud platform, such as one account of the enterprises various pieces for being provided with mixed cloud.Each portion
Divide according to current task, the fortune resource needed for possibility can change.In the present embodiment in order to more
The good unified management for realizing mixed cloud medium cloud resource, while rational allocation private clound domain and public cloud domain include
Fortune resource, in the present embodiment when the resource requirement in the first cloud domain can not by the cloud resource in this cloud domain come
When meeting, at least part of of the slack resources in another cloud domain can again be adjusted by way of dynamic adaptation
With and repartition, so as to meet the cloud domain for having resource requirement.Obviously like this, can not only realize mixing
Dynamic adaptation and the management in the private clound domain and public cloud domain Liang Ge domains medium cloud resource in cloud are closed, and can be carried
The effective rate of utilization of high cloud resource.
To specifically be may include in the resource redistribution in the second cloud domain to the first cloud domain in step s 130：
According to the distributable cloud resource of resource requirement and the first cloud domain, the resource of redistribution is treated in calculating；
Determine the second available slack resources in cloud domain；
Based on the resource for treating redistribution and the slack resources, it is determined that allotment resource.
May include to calculate the resource type and resource quantity of redistribution in the resource for calculating the redistribution
Deng.Certainly determine that slack resources also include determining the type and number of resources of the second available slack resources in cloud domain
Amount etc..
As shown in Fig. 2 used as further improvement of this embodiment, methods described also includes：
Step S140：Determine the priority parameters of the resource requirement；
Step S150：Based on the priority parameters, corresponding account is given by the resource allocation in the first cloud domain
Family, to meet the resource requirement.
Different resource requirements has different urgencies and service request, in the present embodiment with priority parameters come
Embody.The corresponding account level of account such as different in public cloud domain is different.Usual rank account higher,
The grade of service that it is required is higher, and corresponding priority parameters are also higher.
The step of described in the present embodiment, S140 and step S150 can start after the step S130
Complete, or started before the step S130 perform, it is also possible to opened in the step S130
Begin to perform.Such as described first cloud domain receives multiple capacity expansion applications simultaneously.It is surplus that first cloud domain is not used by
Remaining cloud resource, can meet the resource requirement in the capacity expansion application of part, in order to improve response speed, then now
Before performing step S130 just can in advance to the first cloud domain in the present remaining cloud resource for possessing adjust
Match somebody with somebody, to realize that cloud resource allocation is responded as soon as possible.Priority ginseng is further defined in step S140 in this application
Number, usual priority parameters are higher, then it represents that the urgency for meeting the resource requirement is higher, therefore performing
During the step S150, priority parameters resource requirement allotment cloud resource high will be preferably.
Specifically, step S140 includes as described：
Calculate the application weights of the resource requirement；
Application time based on the resource requirement, the application weights and preset function relation, calculate described
Priority parameters.
The application weights can need to meet the emergency of the resource requirement from the dimensional representation of non-temporal.
The application time is the time for proposing the capacity expansion application including the resource requirement, it is also possible to resource pipe
Application time of the reason device according to the capacity expansion application of self-forming after condition detection to resource requirement.When generally applying
Between more early, then satisfaction that should be more preferential, the under normal circumstances application time and the priority parameters
Into positive correlation.
The preset function relation, can be calculated with the application weights and the application time as dependent variable
The corresponding priority parameters of resource requirement.Like this, priority is from time dimension and non-temporal dimensional representation
Meet the emergency of the resource requirement.
During concrete implementation, if although some capacity expansion application time it is later, although application weights compared with
It is small, but with regard to the time-division be promptly it is special the need for carry out the urgent resource requirement of priority adjustment, in this implementation
In example, for also introducing regulatory factor in this urgent resource requirement the present embodiment, by increase regulation because
The priority parameters of the urgent resource requirement of son increase.A kind of situation is there certainly exist, capacity expansion application itself is full of
Pre- first to file, is in no hurry to use, and can also reduce described preferential this when by the regulatory factor
Level parameter.
The functional relation has many kinds, a kind of specific example presented below：
Priorityn=XSLAn*timen*weightn
Wherein, the PrioritynIt is the priority parameters of n-th resource requirement；The weightnIt is n-th
The application weights of individual resource requirement；The timenIt is the application time of n-th resource requirement；Institute
State XSLAnIt is the regulatory factor of n-th resource requirement.The regulatory factor can be positive number.
The application weights for calculating the resource requirement described in the present embodiment, including：
Based on the resource type parameter needed for the corresponding account level of the resource requirement, the resource requirement and
Resource quantity parameter needed for the resource requirement at least one, calculate the application weights.
The usual account level is higher, then mixed cloud needs the emergency for meeting its resource requirement also higher.
The generation account of n-th resource requirement described in aforementioned functions is SLAn。
Cloud resource type needed for the resource requirement can be divided into database resource, middleware resource and outside should
Use resource.
For example when resource type is the database resource, in the resource type parameter for calculating the application weights
May include：What the recovery time target component RTO of the resource requirement requirement, the resource requirement were required answers
Origin target component RPO and backup parameter etc..When resource type is the database resource, for calculating
The resource quantity parameter for stating application weights may include：Memory capacity and memory size etc..
For example when resource type is the middleware, can be wrapped in the resource type parameter for calculating the application weights
Include：The resource requirement requirement thread pool parameter.It is described for calculating when resource type is the middleware
Applying for the resource quantity parameter of weights may include：The parameters such as thread number of concurrent.
In order to avoid the cloud domain of cloud resource ownership simultaneously instead changes repeatedly, it has not been convenient to be managed collectively, and simultaneously
In order to lift resource allocation strategy.The step S110 may include in the present embodiment：The time is specified first
The interior resource requirement for obtaining account.The step S120 includes：Specified in the time second, based on each institute
The resource requirement of account is stated, judges whether the first cloud domain provides enough resources；Described second refers to
Fix time and be later than for the described second specified time.
Like this, specified in the time first and only obtain resource requirement, such as only receive capacity expansion application, temporarily
It is not responding to capacity expansion application.Start to be uniformly processed in the second specified time and resource requirement is obtained within the very first time.
Specific such as described first specified time span for 24 hours, after reaching 24 hours, proceeded by fortune money
Source allotment meets the resource requirement.When satisfaction is over the resource requirement, then there is beginning one to obtain described
The cycle of resource requirement.
In the present embodiment in order to simply dynamically realize the cloud resource between private clound domain and public cloud domain
Allotment, be in the present embodiment based on PAAS resources and SAAS Resource allocation and smoothings.PAAS resources and
After SAAS resources have carried out the division in ownership cloud domain, can directly use, and without reinstalling program etc.
Operation, it becomes possible to realize easy cloud resource allocation.
Specifically, the step S130 may include：
Count the second cloud domain idle PAAS resources and/or SAAS resources；
Based on the resource requirement, the idle PAAS resources and PAAS resources are allocated to described
One cloud domain.
When carrying out the PAAS resources and SAAS allotments in the present embodiment, if private clound domain and public affairs
There is cloud domain to belong to different VLANs, then can be by changing what the network protocol IP address of resource was belonged to
VLAN is realized.If private clound domain and public cloud domain are by accesses control list (Access
Control List, ACL) private clound domain described in control realization and public cloud domain, then can be described by modification
Accesses control list is realized.
Cloud Resource allocation and smoothing method described in the present embodiment, the resource that can enter to rack in mixed cloud domain first unification
Management and dynamic adaptation, such limitation for having broken static allotment in the prior art, meet private clound
The dynamic resource requirement in domain and public cloud domain, while improve the effective rate of utilization of fortune resource, improves mixing
The disposed of in its entirety benefit of cloud platform.
Below in conjunction with above method embodiment, there is provided the device embodiment of above method embodiment can be realized.
Device embodiment：
As shown in figure 3, the present embodiment provides a kind of cloud Resource allocation and smoothing device, described device is applied to include
In the cloud platform in the first cloud domain and the second cloud domain, described device includes：
Acquiring unit 110, the resource requirement of the account for obtaining the first cloud domain；
Judging unit 120, for based on the resource requirement, judging whether the first cloud domain provides enough
Resource；
Deployment unit 130, if enough resources can not be provided for the first cloud domain, based on the money
Source demand is at least partly reassigned to the first cloud domain by the slack resources in the second cloud domain；
Wherein, the first cloud domain is private clound domain, and the second cloud domain is public cloud domain；Or, described
One cloud domain is public cloud domain, and the second cloud domain is private clound domain.
Cloud Resource allocation and smoothing device described in the present embodiment, can be that one or more cloud management equipment or cloud management take
Business device etc..The concrete structure of the acquiring unit 110 may include communication interface, receive the transmission of each account
The monitoring data that capacity expansion application or monitoring device including the resource requirement send.The acquiring unit 110
Processor or process circuit are may also include, can be used to obtain the money by way of analyzing the monitoring data
Source demand.
The concrete structure of the judging unit 120 and the deployment unit 130 may include processor or treatment electricity
Road.The processor or process circuit can be performed by code, realize that above-mentioned judging unit 120 and allotment are single
The function of the analysis resource requirement of the function and the acquiring unit 110 of unit 130.
The processor may include central processor CPU, Micro-processor MCV, digital signal processor or
Programmable array PLC etc..The process circuit may include application-specific integrated circuit ASIC etc..
Device described in the present embodiment can carry out unifying pipe in a word to the cloud resource in private clound domain and public cloud domain
Reason and allotment, can enter the dynamic adaptation of resource that racks according to two kinds of resource requirements in cloud domain, improve cloud money
The utilization ratio in source.
Described device also includes：
Determining unit, the priority parameters for determining the resource requirement；
The deployment unit 130, is additionally operable to based on the priority parameters, by the resource in the first cloud domain
Corresponding account is deployed to, to meet the resource requirement.
Determining unit concrete structure described in the present embodiment may also comprise above-mentioned processor or process circuit or have
Calculator of computing function etc..By the input of variable, the priority is calculated according to preset function relation
Parameter.The multiple resources of satisfaction can be determined in the deployment unit 130 of the present embodiment according to priority parameters
The sequencing of demand, meets each resource requirement successively；It is achieved thereby that meeting urgency resource need high
Preferential satisfaction is sought, so that the users' satisfaction degree of lifting.
The determining unit, the application weights for calculating the resource requirement；Based on the resource requirement
Application time, the application weights and preset function relation, calculate the priority parameters.
Determining unit described in the present embodiment includes calculating the calculator of the application weights, by reading when described
The parameters such as the application time of resource requirement, and calculated based on preset function relation.Described in the present embodiment
Preset function relation may refer to preceding method embodiment.
The determining unit is specific with based on the corresponding account level of the resource requirement, the resource requirement
Resource quantity parameter needed for required resource type parameter and the resource requirement at least one, calculate
The application weights.Herein, the variable that the determining unit calculates the application weights is further disclosed,
These variables can accurately show the emergency of resource requirement, so as to lift user satisfaction.
Additionally, the acquiring unit 110, the resource specifically for specifying acquisition account in the time first need to
Ask；The judging unit 120, specifically for being specified in the time second, the institute based on account each described
Resource requirement is stated, judges whether the first cloud domain provides enough resources；Described second specifies the time to be later than
Described second specifies the time.
Described first specify the time not carry out the response of resource requirement in the present embodiment, i.e., do not enter resource of racking
Allotment, only obtain the resource requirement, second specify the time unify to respond and focus on, this
If sample, treatment effeciency can be lifted.
The deployment unit 130, specifically for count the idle PAAS resources in the second cloud domain and/or
SAAS resources；Based on the resource requirement, the idle PAAS resources and SAAS resources are divided
To the first cloud domain.Deployment unit 130 described in the present embodiment includes counter or with computing function
Processor or process circuit, the PAAS for being available for allotment left unused in the second cloud domain is determined by counting
Resource and/or SAAS resources.
The deployment unit 130 is directly to carry out PAAS resources and SAAS resource allocations in the present embodiment
Result, carrying out PAAS resources and SAAS resources can be by repartitioning PAAS resources and SAAS
The VLAN of resource ownership updates ACL and realizes；The characteristics of with realizing simple.
Below in conjunction with above-mentioned any embodiment, there is provided several specific examples.
Example one：
It is illustrated in figure 4 the mixed cloud that the method described in the embodiment of the present invention can be applied.In Fig. 4 institutes
The mixed cloud for showing includes resource control device, public cloud domain storage resource, private clound domain storage resource, PAAS
Cluster 11, PAAS clusters 12, virtual machine control centre vCenter, PAAS cluster 11, PAAS clusters
12 and gateway device at different levels.Here gateway devices at different levels are including router etc..
Resource management apparatus in the mixed cloud shown in Fig. 4, mixed cloud are changed and have been divided into private clound domain and public affairs
There is cloud domain.The private clound domain includes PAAS clusters 11, PAAS clusters 12 and vCenter at present；Institute
State public cloud domain includes PAAS clusters 1 and PAAS clusters 2 at present.Public cloud domain belongs to and is virtually based on
Net VLAN2, private clound domain belongs to virtual LAN VLAN 1.Set between VLAN1 and VLAN2
It is equipped with fire wall.Private clound domain and public cloud domain by gateway device can realize service convergence and be connected to
Core net, inline isolated area DMZ, outreach DMZ, core production district, customer service system, procotol IP
Private network, intrusion detection IDS equipment, Data Communication Network, China Internet CMNET.Certainly mixing
Cloud finally and external network, when such as CMNET is connected, will be connected by fire wall with exterior gateway.
The specific attachment structure of the mixed cloud of this example may refer to diagram, just no longer detailed one by one again
Describe.
Mixed cloud described in Fig. 4 can apply preceding method embodiment to realize the cloud in private clound domain and public cloud domain
Dynamic resource allocation and management.Resource control device in Fig. 4 can be used as the cloud described in apparatus embodiments
Resource allocation and smoothing device.
As can be seen from Figure 4, in this example using the Model construction mixed cloud in point domain, private clound domain, public cloud
Security boundary is divided by strict VLAN and ACL access controls between domain, resource control device.For
Meet the dynamic adaptation of mixing cloud computing resources, private clound domain, public cloud domain host resource cut-in convergent to system
One service convergence switch network.Fire wall is disposed between PAAS intelligence management and control and public cloud domain, is carried out
Port isolation is controlled with two-way access.
In the mixed cloud shown in Fig. 4, public cloud domain and private clound domain are owned by logically independent PAAS moneys
Source pond, newly applies or during dilatation PAAS resources as user, is preferentially obtained from public cloud domain, if currently
To meet during demand, triggering is provided remaining cloud inadequate resource from resource intelligent control device to private clound domain application cloud
Source.The resource analysis module in private clound domain calculates available minimum PAAS Resource Units, by intelligently managing
Control module adds the unit in public cloud domain, and public cloud domain is by PAAS resource adaptations module to various flat
Taiwan investment source carries out dynamic reorganization, realizes two resource allocations of resource domains.Here minimum PAAS resources list
Unit is to carry out a least amount of resources for cloud scheduling of resource.Herein, this concocting method is equally applicable to privately owned
Resource is allocated in cloud domain from public cloud domain；It is of course also apply to the SAAS moneys between public cloud domain and private clound domain
Source is allocated.
Example two：
As shown in figure 5, this example may be based on the dynamic adaptation side of the PAAS resources of the mixing shown in Fig. 4
Method, specifically includes：
Accounts at different levels apply for PAAS resources in the case where user indicates；The application of PAAS resources is sent to PAAS
Resource control device.PAAS resource control devices, after receiving application, carry out the PAAS in public cloud domain first
Resource allocation.It was found that the PAAS resources in public cloud domain cannot meet the resource requirement of application, to private clound domain
Application PAAS resources.If private clound domain can provide unappropriated PAAS resources, by former private clound
The PAAS resource allocations in domain are in public cloud domain.Allotment is finally based on, by the PAAS resources in private clound domain
In moving to public cloud domain.
Example three：
The specific allotment flow to can be applied to cloud resource of the invention shown in Fig. 6, including：
Step S1：Resource control device receives the n task of resource bid；The n is not less than 1
Integer；
Step S2：Calculate the priority parameters of each task.
Step S3：N task parameters are ranked up to n task according to priority parameters；
Step S4：Based on the process task one by one that sorts；Priority parameters priority treatment high.
Step S5：Return to result.
Based on step S1 to step S5, it is introduced in detailed below.
When user in public cloud domain initiates new application or PAAS resource bids, born by resource control device
Duty resource allocation.Resource control device is using a kind of weighting sequential based on SLA (service-level agreement) point
With strategy.The allocation strategy is described as follows：
1. it is raising allotment efficiency, lifting system utilization of resources quality, resource control device Timing Processing resource
The request of application, i.e., carry out a PAAS resource allocation every T time, between intervals interior receiving
PAAS resource bids are asked, and are not processed.According to production experience, T can be set to 24h.The T
It can be the first specified time in previous embodiment.
2. user's mixing cloud system when PAAS resource bids are initiated is set according to factors such as account levels first
SLA (SLA is the natural number more than or equal to 1), SLA is bigger, represents its grade of service higher；
3. when each user initiates PAAS resource bids, resource control device is equivalent to receiving a task
Task, SLA, application time time of the task comprising this PAAS resource bid, application weights weight.
For example：Application to database resource can be according to it to memory space, internal memory, RPO, RTO, backup etc.
Factor calculates its weights；Application to middleware resource can be calculated according to it to factors such as number of concurrent, thread pools
Its weights.Weight is bigger, and the PAAS resource bid requirements for representing the user are high；
4. assume that resource control device receives 1.....n task in predetermined distance time T, is named as
Task1, task2.....taskn, for each task, will be according to its service-level agreement SLA, application time
The factors such as time, resource bid weight w eight calculate its priority priority.Its computational methods is as follows：
For the n application time time1 of task, time2....timen is ranked up from back to front,
By time1, time2....timen corresponds respectively to a natural number N (1=<N<=n) as its sequential
Value, application time more early task, the corresponding N values of its time are bigger；
Priorityn=XSLAn*timen*weightn
X is the natural number more than or equal to 1, is the regulatory factor that priority parameters calculate Priority, same
When criticizing task priority calculating, X palpus values are consistent.The general values of X can be for 2) in current system.
5. according to the result of calculation of Priority, to task1, task2.....taskn according to Priority from big to small
Order carry out rearrangement TASK1, TASK2 ... TASKm (1=<m<=n)；
6. to TASK1, TASK2 ... TASKm carry out sequential processes to resource control device, its processing method
It is as follows：
Can be to carry out one of handling process of task one by one in the step S4 shown in Fig. 7.
PAAS resources are applied for from public cloud domain based on TASK.TASK represent according to priority parameters
Task after sorting from high to low.For example for the resource bid of TASK1, resource control device is looked into first
Whether ask in its public cloud domain has enough resources to be available for distribution, if public cloud domain resource meets demand directly from
PAAS resources are distributed in public cloud；Transfer treatment if public cloud domain resource is unsatisfactory for the demand of TASK1
TASK2, by that analogy until task is disposed or all tasks finish public cloud domain resource poll；
Judge whether to apply successfully, if applying successfully directly returning to result, if not applying successfully, sentencing
Whether all tasks of breaking applied for PAAS resources to public cloud domain；It is to enter subsequent operation that judged result is,
If judged result is no, next task is obtained according to priority parameters.
Apply for PAAS to private clound domain from for the remaining task order of PAAS resources applied removes task
Resource, and record application result.
Judge all tasks whether to private clound with applied for PAAS resources, if judged result is no, return
Previous action institute is gone back to, result is if the determination result is YES returned.
In above-mentioned processing procedure, task has been disposed, resource control device then directly return all into
Work(treatment mark；If task is also untreated complete, illustrate that public cloud domain resource can not meet the resource of all tasks
Demand.Resource control device is descending one by one to private according to the sequence of priority parameters for remaining task
There is cloud domain to propose PAAS resource bids.After private clound domain is connected to the resource bid request of resource control device,
Available minimum PAAS units are calculated by its resource analysis module first, then by resource control device
By in unit addition public cloud domain PAAS domains.After private clound domain receives total domain PAAS applies,
Resource analysis module calculates PAAS units (such as database, the centre of minimum according to private clound domain resource situation
The virtual machine or physical machine of part platform), resource control device is divided into public affairs by changing the IP of the resource
There is the VLAN in cloud domain, or modification ACL controlling networks are accessed, and realize public cloud domain network up to the PAAS
Resource.PAAS resource adaptations module in the resource control device can be by the new resource dynamic adaptation for adding
To in the PAAS resources of public cloud domain, database, middleware, caching, the integration environment of various patterns are supported
Deng service.
Example four：
(1), when PAAS public clouds domain receives the database (Date Base, DB) transferred in private clound domain
After resource, intelligent management and control module will determine that the DB types at demand end, be started according to different type corresponding suitable
With algorithm, the DB to there is dilatation demand implements dynamic capacity-expanding.
(2), as shown in figure 8, be deployed with the host P1 in public cloud domain three sets of data storehouses (DB1, DB2,
DB3), DB3 requires dilatation CPU, data buffer storage, backup space and changes parameter, but P1 hosts
Without additional hardware resources, modification parameter will influence DB1 and DB2, now using the resource to private clound domain
Draw the dilatation for implementing DB3 by host P2 hosts.
(3) when, implementing dilatation and parameter modification to the DB3 under P1 using P2, intelligent adaptation algorithm is by root
The disturbance degree Y1 of resettlement DB3 is calculated according to the operation factor in three sets of data storehouses, and moves DB1, DB2
Disturbance degree Y2.
When computationally stating disturbance degree, it will use the parameter in table 1 to table 2.
| Data capacity | Operation factor F1 |
| 0-100G | 1 |
| 100G-500G | 2 |
| 500G-1T | 3 |
| 1T-5T | 4 |
| More than 5t | 5 |
Table 1
| Concurrent connection number | Operation factor F2 |
| 0-10 | 1 |
| 10-100 | 2 |
| 100-300 | 3 |
| More than 300 | 4 |
Table 2
| Business continuance | Operation factor F3 |
| Backup and Restore | 1 |
| It is emergent | 2 |
| Dual-active | 3 |
Table 3
| Assurance level | Operation factor F4 |
| 5*8 | 1 |
| 7*24 | 2 |
DB migrates disturbance degree Y1=F1+F2+F3+F4 ..., moves the disturbance degree Y2=of DB1, DB2
F1 '+F2 '+F3 '+F4 ' ..., work as Y1<=Y2, illustrates that resettlement DB3 influences are small, using DB3 is moved, instead
As the same
(4), before DB3 is moved, needed for P2 host's DB main frame machines have reserved DB3 dilatations
Data buffer storage, cpu resource, and parameter modification is completed, 1 grade and 0 DBMS are passed through by resource control device
Backup and Restore, real time data synchronization etc. realize the Data Migration of DB3, and application program is reconnected can be completed
The dilatation of DB3
(5) Y1, is worked as>Y2, implements DB1, DB2 and moves to host P2, the same step (4) of migration scheme,
Remaining resource will supply DB3 after migration, and implement parameter modification, and application program is reconnected can be completed
The dilatation of DB3.
The dilatation of the DB3 specifically can as shown in figure 9, including：
Judge that P2 is by there are available resources, if into next step, the resource in private clound domain is applied for if not；Here
Resource can be database resource etc. in PAAS resources.
Calculate disturbance degree Y1 and disturbance degree Y2.
Judge Y1 whether less than or equal to Y2.
Y1 is less than or equal to Y2, and that implements DB3 moves to P2.
Y1 is more than Y2, implements DB1 and moves to P2.
In several embodiments provided herein, it should be understood that disclosed apparatus and method,
Can realize by another way.Apparatus embodiments described above are only schematical, for example,
The division of the unit, only a kind of division of logic function, can there is other division when actually realizing
Mode, such as：Multiple units or component can be combined, or be desirably integrated into another system, or some are special
Levying to ignore, or does not perform.In addition, the coupling each other of shown or discussed each part,
Or direct-coupling or communication connection can be the INDIRECT COUPLINGs or logical of equipment or unit by some interfaces
Letter connection, can be electrical, machinery or other forms.
It is above-mentioned as separating component illustrate unit can be or may not be it is physically separate, make
For the part that unit shows can be or may not be physical location, you can with positioned at a place,
Can also be distributed on multiple NEs；Can according to the actual needs select therein part or all of
Unit realizes the purpose of this embodiment scheme.
In addition, each functional unit in various embodiments of the present invention can be fully integrated into a processing module
In, or each unit individually as a unit, it is also possible to two or more unit collection
Into in a unit；Above-mentioned integrated unit can both be realized in the form of hardware, it would however also be possible to employ
Hardware adds the form of SFU software functional unit to realize.
One of ordinary skill in the art will appreciate that：Realize all or part of step of above method embodiment
Can be completed by the related hardware of programmed instruction, foregoing program can be stored in a computer-readable
Take in storage medium, the program upon execution, performs the step of including above method embodiment；And it is foregoing
Storage medium include：Movable storage device, read-only storage (ROM, Read-Only Memory),
Random access memory (RAM, Random Access Memory), magnetic disc or CD etc. are various
Can be with the medium of store program codes.
The above, specific embodiment only of the invention, but protection scope of the present invention do not limit to
In this, any one skilled in the art the invention discloses technical scope in, can be easily
Expect change or replacement, should all be included within the scope of the present invention.Therefore, protection of the invention
Scope should be based on the protection scope of the described claims.
Claims (12)
1. a kind of cloud Resource allocation and smoothing method, it is characterised in that methods described be applied to include the first cloud domain and
In the cloud platform in the second cloud domain, methods described includes：
Obtain the resource requirement of the account in the first cloud domain；
Based on the resource requirement, judge whether the first cloud domain provides enough resources；
If the first cloud domain can not provide enough resources, based on the resource requirement by second cloud
The slack resources in domain are at least partly reassigned to the first cloud domain；
Wherein, the first cloud domain is private clound domain, and the second cloud domain is public cloud domain；Or, described
One cloud domain is public cloud domain, and the second cloud domain is private clound domain.
2. method according to claim 1, it is characterised in that
Methods described also includes：
Determine the priority parameters of the resource requirement；
Based on the priority parameters, corresponding account is given by the resource allocation in the first cloud domain, to meet
The resource requirement.
3. method according to claim 2, it is characterised in that
The priority parameters for determining the resource requirement, including：
Calculate the application weights of the resource requirement；
Application time based on the resource requirement, the application weights and preset function relation, calculate described
Priority parameters.
4. method according to claim 3, it is characterised in that
The application weights for calculating the resource requirement, including：
Based on the resource type parameter needed for the corresponding account level of the resource requirement, the resource requirement and
Resource quantity parameter needed for the resource requirement at least one, calculate the application weights.
5. method according to claim 1, it is characterised in that
The resource requirement of the account for obtaining the first cloud domain, including：
In the first resource requirement for specifying acquisition account in the time；
It is described to judge whether the first cloud domain provides enough resources based on the resource requirement, including：
Specified in the time second, based on the resource requirement of account each described, judge first cloud
Whether domain provides enough resources；
Described second specifies the time to be later than described second specifies the time.
6. method according to claim 1, it is characterised in that
If the first cloud domain can not provide enough resources, based on the resource requirement by described
The slack resources in two cloud domains are at least partly reassigned to the first cloud domain, including：
Count the second cloud domain idle PAAS resources and/or SAAS resources；
Based on the resource requirement, the idle PAAS resources and SAAS resources are allocated to described
One cloud domain.
7. a kind of cloud Resource allocation and smoothing device, it is characterised in that described device be applied to include the first cloud domain and
In the cloud platform in the second cloud domain, described device includes：
Acquiring unit, the resource requirement of the account for obtaining the first cloud domain；
Judging unit, for based on the resource requirement, judging whether the first cloud domain provides enough money
Source；
Deployment unit, if can not provide enough resources for the first cloud domain, is needed based on the resource
Ask and the slack resources in the second cloud domain are at least partly reassigned to the first cloud domain；
Wherein, the first cloud domain is private clound domain, and the second cloud domain is public cloud domain；Or, described
One cloud domain is public cloud domain, and the second cloud domain is private clound domain.
8. device according to claim 7, it is characterised in that
Described device also includes：
Determining unit, the priority parameters for determining the resource requirement；
The deployment unit, is additionally operable to based on the priority parameters, by the resource allocation in the first cloud domain
To corresponding account, to meet the resource requirement.
9. device according to claim 8, it is characterised in that
The determining unit, the application weights for calculating the resource requirement；Based on the resource requirement
Application time, the application weights and preset function relation, calculate the priority parameters.
10. device according to claim 9, it is characterised in that
The determining unit is specific with based on the corresponding account level of the resource requirement, the resource requirement
Resource quantity parameter needed for required resource type parameter and the resource requirement at least one, calculate
The application weights.
11. devices according to claim 7, it is characterised in that
The acquiring unit, specifically for specifying the resource requirement that account is obtained in the time first；
The judging unit, specifically for being specified in the time second, the money based on account each described
Source demand, judges whether the first cloud domain provides enough resources；Described second specified time was later than described
Second specifies the time.
12. devices according to claim 7, it is characterised in that
The deployment unit, specifically for counting the idle PAAS resources in the second cloud domain and/or SAAS
Resource；
Based on the resource requirement, the idle PAAS resources and SAAS resources are allocated to described
One cloud domain.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510430509.4A CN106713375A (en) | 2015-07-21 | 2015-07-21 | Method and device for allocating cloud resources |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510430509.4A CN106713375A (en) | 2015-07-21 | 2015-07-21 | Method and device for allocating cloud resources |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106713375A true CN106713375A (en) | 2017-05-24 |
Family
ID=58899929
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510430509.4A Pending CN106713375A (en) | 2015-07-21 | 2015-07-21 | Method and device for allocating cloud resources |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106713375A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107526818A (en) * | 2017-08-28 | 2017-12-29 | 中山市博美科研技术服务有限公司 | One kind is based on enterprise's private clound big data processing system and method |
| CN108306981A (en) * | 2018-03-08 | 2018-07-20 | 江苏省广播电视总台 | The dual isomery PaaS redundancy approach of media cloud platform service layer safety can be enhanced |
| CN108551399A (en) * | 2018-03-29 | 2018-09-18 | 深信服科技股份有限公司 | Dispositions method, system and the relevant apparatus serviced under a kind of cloud environment |
| CN109032788A (en) * | 2018-06-29 | 2018-12-18 | 北京百度网讯科技有限公司 | Reserved resource pool dynamic dispatching method, device, computer equipment and storage medium |
| CN109120702A (en) * | 2018-08-24 | 2019-01-01 | 高振宇 | The method and system of cloud are isolated |
| CN109936592A (en) * | 2017-12-15 | 2019-06-25 | 中国移动通信集团浙江有限公司 | Method, apparatus, electronic equipment and the storage medium of service processing |
| CN112565317A (en) * | 2019-09-25 | 2021-03-26 | 阿里巴巴集团控股有限公司 | Hybrid cloud system, data processing method and device thereof, and storage medium |
| CN114155969A (en) * | 2021-11-16 | 2022-03-08 | 医渡云（北京）技术有限公司 | Medical data acquisition method and device, electronic equipment and storage medium |
| US20230112031A1 (en) * | 2021-10-09 | 2023-04-13 | Dell Products L.P. | System and method for workload management in a distributed system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102195956A (en) * | 2010-03-19 | 2011-09-21 | 富士通株式会社 | Cloud service system and user right management method thereof |
| CN102292698A (en) * | 2009-02-04 | 2011-12-21 | 思杰系统有限公司 | Methods and systems for automated management of virtual resources in a cloud computing environment |
| CN102427473A (en) * | 2011-11-28 | 2012-04-25 | 中国联合网络通信集团有限公司 | Method and system for constructing cross-platform resource |
| CN102594869A (en) * | 2011-12-30 | 2012-07-18 | 深圳市同洲视讯传媒有限公司 | Method and device for dynamically distributing resources under cloud computing environment |
| CN103430150A (en) * | 2010-11-22 | 2013-12-04 | 瑞典爱立信有限公司 | Technique for resource creation in a cloud computing system |
| CN103793656A (en) * | 2012-10-30 | 2014-05-14 | 三星Sds株式会社 | Security implemented through metadata orchestrators |
-
2015
- 2015-07-21 CN CN201510430509.4A patent/CN106713375A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102292698A (en) * | 2009-02-04 | 2011-12-21 | 思杰系统有限公司 | Methods and systems for automated management of virtual resources in a cloud computing environment |
| CN102195956A (en) * | 2010-03-19 | 2011-09-21 | 富士通株式会社 | Cloud service system and user right management method thereof |
| CN103430150A (en) * | 2010-11-22 | 2013-12-04 | 瑞典爱立信有限公司 | Technique for resource creation in a cloud computing system |
| CN102427473A (en) * | 2011-11-28 | 2012-04-25 | 中国联合网络通信集团有限公司 | Method and system for constructing cross-platform resource |
| CN102594869A (en) * | 2011-12-30 | 2012-07-18 | 深圳市同洲视讯传媒有限公司 | Method and device for dynamically distributing resources under cloud computing environment |
| CN103793656A (en) * | 2012-10-30 | 2014-05-14 | 三星Sds株式会社 | Security implemented through metadata orchestrators |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107526818A (en) * | 2017-08-28 | 2017-12-29 | 中山市博美科研技术服务有限公司 | One kind is based on enterprise's private clound big data processing system and method |
| CN109936592A (en) * | 2017-12-15 | 2019-06-25 | 中国移动通信集团浙江有限公司 | Method, apparatus, electronic equipment and the storage medium of service processing |
| CN108306981A (en) * | 2018-03-08 | 2018-07-20 | 江苏省广播电视总台 | The dual isomery PaaS redundancy approach of media cloud platform service layer safety can be enhanced |
| CN108551399A (en) * | 2018-03-29 | 2018-09-18 | 深信服科技股份有限公司 | Dispositions method, system and the relevant apparatus serviced under a kind of cloud environment |
| CN109032788A (en) * | 2018-06-29 | 2018-12-18 | 北京百度网讯科技有限公司 | Reserved resource pool dynamic dispatching method, device, computer equipment and storage medium |
| CN109120702A (en) * | 2018-08-24 | 2019-01-01 | 高振宇 | The method and system of cloud are isolated |
| CN109120702B (en) * | 2018-08-24 | 2021-07-27 | 高振宇 | Method and system for isolating cloud |
| CN112565317A (en) * | 2019-09-25 | 2021-03-26 | 阿里巴巴集团控股有限公司 | Hybrid cloud system, data processing method and device thereof, and storage medium |
| CN112565317B (en) * | 2019-09-25 | 2022-04-08 | 阿里巴巴集团控股有限公司 | Hybrid cloud system, data processing method and device thereof, and storage medium |
| US20230112031A1 (en) * | 2021-10-09 | 2023-04-13 | Dell Products L.P. | System and method for workload management in a distributed system |
| CN114155969A (en) * | 2021-11-16 | 2022-03-08 | 医渡云（北京）技术有限公司 | Medical data acquisition method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106713375A (en) | Method and device for allocating cloud resources | |
| CN108829494B (en) | Container cloud platform intelligent resource optimization method based on load prediction | |
| CN105843670B (en) | A kind of cloud platform virtual cluster deployment integration method | |
| CN104881325B (en) | A kind of resource regulating method and resource scheduling system | |
| CN111966500B (en) | Resource scheduling method and device, electronic equipment and storage medium | |
| CN103870314B (en) | Method and system for simultaneously operating different types of virtual machines by single node | |
| CN103248659B (en) | A kind of cloud computing resource scheduling method and system | |
| CN104102543B (en) | The method and apparatus of adjustment of load in a kind of cloud computing environment | |
| CN110661842B (en) | Resource scheduling management method, electronic equipment and storage medium | |
| CN108614726A (en) | Virtual machine creation method and device | |
| CN109783225B (en) | Tenant priority management method and system of multi-tenant big data platform | |
| CN107968802A (en) | The method, apparatus and filtering type scheduler of a kind of scheduling of resource | |
| CN105893113A (en) | Management system and management method of virtual machine | |
| CN114356543A (en) | Kubernetes-based multi-tenant machine learning task resource scheduling method | |
| CN104239123A (en) | Campus-cloud-platform-oriented virtual machine management dispatching method and system | |
| CN107346264A (en) | A kind of method, apparatus and server apparatus of virtual machine load balance scheduling | |
| CN106385329A (en) | Processing method and device of resource pool and equipment | |
| CN110187960A (en) | A kind of distributed resource scheduling method and device | |
| CN104301257B (en) | A kind of resource allocation methods, device and equipment | |
| CN103701934A (en) | Resource optimal scheduling method and virtual machine host machine optimal selection method | |
| CN110196773B (en) | Multi-time-scale security check system and method for unified scheduling computing resources | |
| CN107864211A (en) | Cluster resource dispatching method and system | |
| CN109522090A (en) | Resource regulating method and device | |
| CN114968601B (en) | Scheduling method and scheduling system for AI training jobs with resources reserved in proportion | |
| CN115658311A (en) | Resource scheduling method, device, equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170524 |