CN103503404A - Resource scheduling method, device and system - Google Patents
Resource scheduling method, device and system Download PDFInfo
- Publication number
- CN103503404A CN103503404A CN201180003668.6A CN201180003668A CN103503404A CN 103503404 A CN103503404 A CN 103503404A CN 201180003668 A CN201180003668 A CN 201180003668A CN 103503404 A CN103503404 A CN 103503404A
- Authority
- CN
- China
- Prior art keywords
- virtual machine
- application
- cloud platform
- resources
- external
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; 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
- G06F9/5072—Grid computing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Stored Programmes (AREA)
- Debugging And Monitoring (AREA)
Abstract
本发明涉及一种资源调度方法、装置和系统，方法包括：获取应用中当前运行的各进程的业务量，所述应用部署在内部云平台的虚拟机集群中；在所述应用中当前运行的各进程的业务量满足预设条件时，根据预先获取的所述应用的进程伸缩策略，确定所述虚拟机集群需要扩展的资源数；在确定向外部云平台申请调度资源时，根据确定的所述资源数向外部云平台发送虚拟机申请请求；在接收到所述外部云平台返回的虚拟机信息之后，向所述虚拟机信息指向的外部虚拟机部署所述应用的进程。本发明通过调度外部云资源对内部云资源进行补充，从而保障了应用的服务质量。
The present invention relates to a resource scheduling method, device and system. The method includes: acquiring the business volume of each process currently running in an application, the application being deployed in a virtual machine cluster of an internal cloud platform; When the business volume of each process satisfies the preset condition, determine the number of resources that the virtual machine cluster needs to expand according to the pre-acquired process scaling strategy of the application; Send a virtual machine application request to the external cloud platform according to the number of resources; after receiving the virtual machine information returned by the external cloud platform, deploy the application process to the external virtual machine pointed to by the virtual machine information. The present invention supplements internal cloud resources by dispatching external cloud resources, thereby ensuring the service quality of applications.
Description
Resource regulating method, device and system technical field
The present embodiments relate to the network communications technology, more particularly to a kind of resource regulating method, device and system.Background technology
Cloud is a kind of emerging network distributed computing technology, can be packaged specific ins and outs by cloud, then provide service by providing a user relatively simple interface.Many enterprises are all deployed with privately owned cloud platform at present, will apply and move to the respective privately owned cloud platform of enterprise.Such as China Telecom, movement enterprise, their application include enterprise resource planning（Enterprise Resource Planning, abbreviation ERP), charging, online business hall, billing statistics analysis, marketing analysis, portal website etc., respective application is deployed in the privately owned cloud platform of enterprise itself by they.
Generally, the number of resources that enterprise can be according to needed for being calculated the portfolio of concrete application, so as to be the application deployment respective resources in the privately owned cloud platform of enterprise.Prior art in the privately owned cloud platform of enterprise for some application deployment resource, it is by calculating each time zone using required resource, then according to the required maximum using resource of each time zone application, the static configuration of resource is carried out to the cluster virtual machine of internal cloud platform.
But, the portfolio of the application of most enterprises is in different time zone and unstable, for example：The portfolio of the application of most enterprises will be increased sharply in the end of the year, but then relatively more steady in other time zones.If the maximum for the number of resources that enterprise calculates according to each time zone carries out applying deployment, more resource is in idle condition in the relatively low most time zones of portfolio, configured resource, causes the wasting of resources.If enterprise, then can be because of the inadequate resource disposed in advance, so that corresponding service can not be provided when traffic peaks occurs in the application not by the maximum deployment resource of the portfolio calculated using each time zone.The content of the invention
The embodiment of the present invention provides a kind of resource regulating method, device and system, to be supplemented by dispatching outside cloud resource internal cloud resource, so as to ensure the service quality of application.
The embodiments of the invention provide a kind of resource regulating method, including：
The portfolio for each process currently run in application is obtained, the application deployment is internally in the cluster virtual machine of cloud platform；
When the portfolio for each process currently run in the application meets preparatory condition, according to the flexible strategy of the process of the application obtained in advance, determine that the cluster virtual machine needs the number of resources extended；It is determined that during to outside cloud platform application scheduling resource, sending virtual machine application to outside cloud platform according to the number of resources of determination and asking；
After the virtual machine information that the outside cloud platform is returned is received, the external Virtual machine pointed to the virtual machine information disposes the process of the application.
The embodiment of the present invention additionally provides a kind of resource scheduling device, including：
Information obtains ear not block, the portfolio for obtaining each process currently run in application, and the application deployment is internally in the cluster virtual machine of cloud platform；
Number of resources determining module, when the portfolio of each process for currently running in the application meets preparatory condition, according to the flexible strategy of the process of the application obtained in advance, determines that the cluster virtual machine needs the number of resources extended；
Scheduling of resource application module, for it is determined that during to outside cloud platform application scheduling resource, sending virtual machine application to outside cloud platform according to the number of resources of determination and asking；
Application process transferring module, for after the virtual machine information that the outside cloud platform is returned is received, the external Virtual machine pointed to the virtual machine information to dispose the process of the application.
The embodiment of the present invention additionally provides a kind of resource scheduling system, including：
Internal cloud platform and outside cloud platform；And
Above-mentioned resource scheduling device；The resource scheduling device is communicated to connect with the internal cloud platform and the outside cloud platform.
Resource regulating method provided in an embodiment of the present invention, device and system, can carry out the dynamic dispatching of resource by granularity of process according to the portfolio of each process of the application disposed to cluster virtual machine.During the dynamic dispatching of resource, if the idling-resource of internal cloud platform can not meet the business increased requirement of the application, resource supplement can be carried out to outside cloud platform application resource, and carries out the procedure deployment of external Virtual machine, the service quality of application has thus been ensured.Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, the required accompanying drawing used in embodiment or description of the prior art will be briefly described below, apparently, drawings in the following description are only some embodiments of the present invention, for those of ordinary skill in the art
Without having to pay creative labor, other accompanying drawings can also be obtained according to these accompanying drawings.Fig. 1 is the resource regulating method flow chart that the embodiment of the present invention one is provided；
Fig. 2 is the structural representation for the resource scheduling system that the embodiment of the present invention two is provided；
Fig. 3 is the internal structure example of internal cloud platform in Fig. 2；
Fig. 4 is the signaling interaction diagram for the resource regulating method that the embodiment of the present invention three is provided；
Fig. 5 is the signaling interaction diagram for the resource regulating method that the embodiment of the present invention four is provided；
Fig. 6 is the structural representation for the resource scheduling device that the embodiment of the present invention five is provided.Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.Based on the embodiment in the present invention, the every other embodiment that those of ordinary skill in the art are obtained under the premise of creative work is not paid belongs to the scope of protection of the invention.
The sequence number of following examples of the present invention is for illustration only, and the quality of embodiment is not represented.
Fig. 1 is the resource regulating method flow chart that the embodiment of the present invention one is provided.Method as shown in Figure 1 includes：
Step 11:The portfolio for each process currently run in application is obtained, the application deployment is internally in the cluster virtual machine of cloud platform.
Internal cloud platform is deployed with multiple cluster virtual machines in advance；One cluster virtual machine includes multiple virtual machines；One cluster virtual machine can dispose an application or multiple applications；One application may include one or more processes；One process can be deployed on one or more virtual machines.
After internally the application deployment of cloud platform is completed, the real time traffic load for the application disposed in internal cloud platform in cluster virtual machine can be obtained.The real time traffic load of certain application may include：The information such as process, the portfolio of each process currently run in the application；User's number of request information that the portfolio of a certain process such as process is received.
Step 12:When the portfolio for each process currently run in the application meets preparatory condition, according to the flexible strategy of the process of the application obtained in advance, determine that the cluster virtual machine needs the number of resources extended.
Preparatory condition described in this step can be set according to actual needs.For example, preparatory condition can be：The portfolio sum of each process of application is more than the first pre-determined threshold, and the idling-resource of cluster virtual machine can not
Meet the portfolio increased requirement for applying each process；Or, preparatory condition can also be：The respective portfolio of one or more processes currently run in is more than the second pre-determined threshold, and the idling-resource of cluster virtual machine can not meet these process portfolio increased requirements.Wherein, the idling-resource of above-mentioned cluster virtual machine can be obtained in advance.
If the portfolio for each process currently run in the application meets preparatory condition, according to the flexible strategy of the process of the application obtained in advance, determine that the cluster virtual machine needs the number of resources extended.
The flexible strategy of the process of the application can be obtained in advance, acquisition in the application deployment request such as provided from user.The flexible strategy of process may include：Increase the application enters number of passes；Under the situation, the number of resources that cluster virtual machine needs extend is：Increase the number of resources needed for the process of predetermined number.Or, the flexible strategy of process may include：Increase the number of resources for the one or more processes currently run in the application；Under the situation, the number of resources that the cluster virtual machine needs extend is：The increased total number resource for needed for each process currently run in the application.
Step 13:It is determined that during to outside cloud platform application scheduling resource, sending virtual machine application to outside cloud platform according to the number of resources of determination and asking.
This step can be according to predetermined policy, and whether decision-making is to outside cloud platform application scheduling resource.
Optionally, this step can be according to the idling-resource situation of internal cloud platform, and decision-making is to outside cloud platform or internal cloud platform application scheduling resource.For example：If the resource of internal cloud platform is more sufficient, the idling-resource of internal cloud platform can provide the scheduling of the number of resources of determination, then internally in the idling-resource of cloud platform, the number of resources determined can be dispatched for the cluster virtual machine.If the inadequate resource of internal cloud platform, the idling-resource of internal cloud platform can not provide the scheduling of the number of resources of determination, virtual machine application then can be sent to outside cloud platform according to the number of resources of determination to ask, for outside cloud platform application scheduling resource.
Optionally, this step can also be according to the portfolio variation tendency of each process currently run in the application, and whether decision-making is to outside cloud platform application scheduling resource.If it is an acts and efforts for expediency that the portfolio for each process currently run in the application, which increases, it is determined that to outside cloud platform application scheduling resource；If it is a long-term action that the portfolio for each process currently run in the application, which increases, it is determined that to outside cloud platform application scheduling resource.
Step 14:After the virtual machine information that the outside cloud platform is returned is received, the external Virtual machine pointed to the virtual machine information disposes the process of the application.
For ease of describing, in the embodiment of the present invention：By the void of the physical resource deployment based on internal cloud platform
Plan machine, referred to as internal virtual machine；By the virtual machine of the physical resource deployment based on outside cloud platform, referred to as external Virtual machine.
Outside cloud platform, can be according to virtual machine application request deployment external Virtual machine after virtual machine application request is received, and returns to virtual machine information；The virtual machine information includes：The information such as the reference address of external Virtual machine., can be by the process of the application, according on default process migration policy deployment to external Virtual machine after the virtual machine information that outside cloud platform is returned is received.
Default process migration strategy is for example：The process of pre-determined threshold will be less than in application with other process interaction frequency, i.e. the first process is disposed the internal virtual machine of the cluster virtual machine of the application from internal cloud platform, moved on external Virtual machine.After on the first process migration to external Virtual machine, internal cloud platform disposes the resource that first process of the cluster virtual machine release of the application takes, available for being supplied to other processes of the application to use, to meet the requirement of the application service quality.Further, since the first process and the frequency of other process interactions are relatively low, therefore, by the first process migration to external Virtual machine, the interaction between too many process will not be increased, thus the influence to the performance and service quality of application is smaller.
Or, presetting process migration strategy can also be such as：All processes currently run in the application are disposed on outside virtual machine, and the partial service load of the application is moved into external Virtual machine from the internal virtual machine of cluster virtual machine.So after deployment, the application can carry out load balancing by internal virtual machine and external Virtual machine, thus ensure the service quality of the application.
Above two process migration strategy can be applied to dispose the situation of single application in a cluster virtual machine, apply also for disposing the situation of multiple applications in a cluster virtual machine.
If in addition, in the case of a cluster virtual machine disposes multiple applications, presetting process migration strategy can also be such as：An application in multiple applications that the cluster virtual machine is disposed, i.e., all processes currently run in the first application and the application, external Virtual machine is all moved to from the internal virtual machine of cluster virtual machine.So after deployment, the resource that first application of cluster virtual machine release takes, provides service, to ensure the service quality of other application available for the other application disposed for the cluster virtual machine, and the business and its service quality of the first application, it can be provided by external Virtual machine.
The resource regulating method that the present embodiment is provided, can be according to the portfolio of each process of the application disposed to cluster virtual machine, the dynamic dispatching of progress resource by granularity of process.During the dynamic dispatching of resource, if the idling-resource of internal cloud platform can not meet the business increased requirement of the application, resource supplement can be carried out to outside cloud platform application resource, and carries out the procedure deployment of external Virtual machine, the service quality of application has thus been ensured.
Fig. 2 is the structural representation for the resource scheduling system that the embodiment of the present invention two is provided.Resource scheduling system as shown in Figure 2 includes：Internal cloud platform 21, resource scheduling device 22 and outside cloud platform 23;Resource scheduling device 22 is communicated to connect with internal cloud platform 21 and outside cloud platform 23 respectively.
The internal structure of internal cloud platform 21 can be found in shown in Fig. 3, specifically, internal cloud platform can the physical resource based on internal cloud platform dispose multiple cluster virtual machines, each cluster virtual machine includes one or more virtual machines, and a cluster virtual machine includes multiple virtual machines；One cluster virtual machine can dispose an application or multiple applications；One application may include to multiple processes；One process can be deployed on one or more virtual machines.Internal cloud platform 21 can be privately owned cloud platform；Or, internal cloud platform 21 can be publicly-owned cloud platform, such as expansible cloud computing of Amazon elasticity（Amazon Elastic Compute Cloud, abbreviation Amazon EC2) platform etc..
When the business load that resource scheduling device 22 is responsible for a certain application that cloud platform has been disposed internally meets the idling-resource of preparatory condition and internal cloud platform and can not meet application service quality, to resources such as outside cloud platform application external Virtual machines.
Outside cloud platform 23 can be privately owned cloud platform；Or, outside cloud platform 23 can be publicly-owned cloud platform, such as Amazon EC2, IBM dilatations on demand（IBM Capacity on Demand, abbreviation IBM COD), the cloud platform such as Hewlett-Packard's business event (HP Enterprise service).
Fig. 4 is the signaling interaction diagram for the resource regulating method that the embodiment of the present invention three is provided.The present embodiment is illustrated by taking the resource scheduling system shown in Fig. 2 and Fig. 3 as an example.Exemplified by one cluster virtual machine of the present embodiment, and the cluster virtual machine disposes an application.As in Figure 2-4, the resource regulating method that the present embodiment is provided includes：
Step 41:Resource scheduling device collects the cluster virtual machine information information related to application service quality.Cluster virtual machine information includes：The information such as the identifying of the cluster virtual machine disposed, application, the resource service condition of each cluster virtual machine of the deployment of each cluster virtual machine.
The related information of application service quality includes：Using information such as corresponding cluster virtual machine, the loading conditions of application；The loading condition of application includes：The information such as process, the portfolio of each process currently run in.
Step 42:Resource scheduling device judges whether the portfolio for each process currently run in the application meets preparatory condition according to the application service quality relevant information of the application, if it is satisfied, performing step 43;Otherwise, step 41 is performed.
Preparatory condition described in this step can be set according to actual needs.For example, preparatory condition can be：Should
The portfolio sum of each process is more than the first pre-determined threshold, and the idling-resource of cluster virtual machine can not meet the portfolio increased requirement for applying each process；Or, preparatory condition can also be：The respective portfolio of one or more processes currently run in is more than the second pre-determined threshold, and the idling-resource of cluster virtual machine can not meet these process portfolio increased requirements.Wherein, the idling-resource of above-mentioned cluster virtual machine can be obtained in advance.
If the portfolio for each process currently run in the application meets preparatory condition, according to the flexible strategy of the process of the application obtained in advance, determine that the cluster virtual machine needs the number of resources extended.
Step 43:Resource scheduling device obtains the flexible strategy of process of the application, determines that cluster virtual machine needs the number of resources extended according to the flexible strategy of the process.
The flexible strategy of the process of the application can be obtained in advance, acquisition in the application deployment request such as provided from user.The flexible strategy of process may include：Increase the application enters number of passes；Under the situation, the number of resources that cluster virtual machine needs extend is：Increase the number of resources needed for the process of predetermined number.Or, the flexible strategy of process may include：Increase the number of resources for the one or more processes currently run in the application；Under the situation, the number of resources that the cluster virtual machine needs extend is：The increased total number resource for needed for each process currently run in the application.
Step 44:Resource scheduling device can be according to the application service quality relevant information of the application, and decision-making is internally cloud platform or outside cloud platform application resource, if it is determined that internally cloud platform application resource, then performs step 45;If it is determined that to outside cloud platform application resource, then performing step 49.
Resource scheduling device can analyze the portfolio variation tendency for applying each process according to the application service quality relevant information of the application, and determine it is internally cloud platform or to outside cloud platform application scheduling resource according to analysis result., can internally cloud platform application resource if analysis result shows that the traffic load of each process of the application is long term phenomenon.If analysis result shows that the traffic load of each process of the application is short-term phenomenon, to outside cloud platform application resource, can thus avoid the occurrence of internal cloud platform physical resource blindly extend caused by resource utilization it is low the problem of.
Step 45:Internally cloud platform sends resource bid request to resource scheduling device, for asking to dispatch the number of resources determined for the cluster virtual machine.
Resource bid request may include：The information such as the number of resources, the mark of cluster virtual machine determined.Step 46:Internal cloud platform receives the resource bid request, it is determined whether can provide the scheduling of the number of resources of determination, if it is, performing step 47;Otherwise step 49 is performed.
Internal cloud platform asks the number of resources of application according to itself idling-resource number and resource bid, it is determined that
Whether this scheduling of resource can be provided.
If the resource of internal cloud platform is more sufficient, the idling-resource of such as internal cloud platform is more than or equal to the number of resources determined, then internal cloud platform can provide the scheduling of the number of resources of determination.
If the inadequate resource of internal cloud platform, the idling-resource of such as internal cloud platform is less than the number of resources determined, then internal cloud platform can not provide the scheduling of the number of resources of determination.
Step 47:Internal cloud platform distributes the physical resource matched with the number of resources determined in itself idling-resource, and the information of the physical resource of distribution is sent into resource scheduling device.
If internal cloud platform is determined to provide the scheduling of the number of resources determined, then internally in the idling-resource of cloud platform, the physical resource matched with the number of resources determined is distributed, and the information of the physical resource of distribution is sent to resource scheduling device.
Step 48:The physical resource that resource scheduling device is distributed based on internal cloud platform, according to the flexible tactful flexible processing of carry out process of above-mentioned process.
The flexible processing of resource scheduling device process is for example：Resource scheduling device increases the process for applying predetermined number newly in cluster virtual machine, or, resource scheduling device, which increases the application, to be needed to increase the governable memory source of one or more processes of resource.
Step 49:Resource scheduling device sends virtual machine application to outside cloud platform and asked according to the number of resources of determination.
Virtual machine application request includes：The specification and quantity of virtual machine, virtual machine need the information such as image banner to be mounted, the corresponding cluster virtual machine information of virtual machine, outside cloud platform mark.
Virtual machine needs the corresponding file of image banner to be mounted, can be uploaded to by the interface of outside cloud platform in outside cloud platform.
Outside cloud platform can be privately owned cloud platform；Or, outside cloud platform can be publicly-owned cloud platform, such as
The cloud platforms such as Amazon EC2, IBM COD, Hewlett-Packard's business event.
Resource scheduling device is specific from which outside cloud platform application resource, can be pre-configured with by keeper.Or, resource scheduling device can be according to predetermined condition automatic decision from which outside cloud platform application resource, such as the specification information for the virtual machine that the outside cloud platform that basis is obtained in advance can be provided, it is possible to provide the required virtual machine specification same size of Current resource scheduling or the outside cloud platform close to specification, is used as target external cloud platform.
Step 410:Outside cloud platform receives virtual machine application request, according to virtual machine application request deployment external Virtual machine in the physical resource of outside cloud platform.Virtual machine information is returned to resource scheduling device.
Step 411：Outside cloud platform returns to virtual machine information to resource scheduling device.
Virtual machine information may include：Reference address of external Virtual machine etc..
Step 412:Resource scheduling device stores the relevant information of external Virtual machine.
The relevant information of resource scheduling device storage external Virtual machine may include：The corresponding cluster virtual machine information of reference address, external Virtual machine, Resource Properties of external Virtual machine of external Virtual machine etc.；Wherein the Resource Properties of external Virtual machine can be：Outside cloud resource.
Step 413:Resource scheduling device is handled according to the flexible tactful carry out process of above-mentioned process is flexible, and on the external Virtual machine that the virtual machine information received is pointed to, according to the process of the default process migration policy deployment application.
The flexible processing of resource scheduling device process is for example：Resource scheduling device increases the process for applying predetermined number newly in cluster virtual machine, or, resource scheduling device, which increases the application, to be needed to increase the governable memory source of one or more processes of resource.
Default process migration strategy is for example：The process of pre-determined threshold will be less than in application with other process interaction frequency, i.e. the first process is disposed the internal virtual machine of the cluster virtual machine of the application from internal cloud platform, moved on external Virtual machine.After on the first process migration to external Virtual machine, internal cloud platform disposes the resource that first process of the cluster virtual machine release of the application takes, available for being supplied to other processes of the application to use, to meet the requirement of the application service quality.Further, since the first process and the frequency of other process interactions are relatively low, therefore, by the first process migration to external Virtual machine, the interaction between too many process will not be increased, thus the influence to the performance and service quality of application is smaller.
Or, presetting process migration strategy can also be such as：All processes currently run in the application are disposed on outside virtual machine, and the partial service load of the application is moved into external Virtual machine from the internal virtual machine of cluster virtual machine.So after deployment, the application can carry out load balancing by internal virtual machine and external Virtual machine, thus ensure the service quality of the application.
The present embodiment is completed in resource scheduling device after this scheduling of resource, repeats above-mentioned steps, the dynamic dispatching of resource can be carried out by granularity of process according to the real time business load of the application disposed to cluster virtual machine.During the dynamic dispatching of resource, it can determine that internally cloud platform, still to outside cloud platform scheduling resource, improves the flexibility of scheduling of resource according to the traffic load trend of application.In addition, when the idling-resource of internal cloud platform can not meet the business increased requirement of the application, resource supplement can be carried out to outside cloud platform application resource, and the procedure deployment of external Virtual machine is carried out using flexible process migration strategy, the service quality of application has thus been ensured.
Fig. 5 is the signaling interaction diagram for the resource regulating method that the embodiment of the present invention four is provided.The present embodiment is illustrated by taking the resource scheduling system shown in Fig. 2 and Fig. 3 as an example.Exemplified by one cluster virtual machine of the present embodiment, and the cluster virtual machine is deployed with multiple applications.As shown in Fig. 2, Fig. 3 and Fig. 5, the resource regulating method that the present embodiment is provided includes：
Step 51:Resource scheduling device collects the cluster virtual machine information information related to the application service quality for each application that cluster virtual machine is disposed.
Step 52:Resource scheduling device judges whether the portfolio for each process currently run in each application meets preparatory condition respectively according to the application service quality relevant information respectively applied in cluster virtual machine, if it is satisfied, performing step 53;Otherwise, step 51 is performed.
Step 53- steps 512:In multiple applications that the application that step 53- steps 512 are related to is disposed for the cluster virtual machine, the portfolio for each process currently run meets the application of preparatory condition, and other technologies scheme is similar to step 43- steps 412, will not be repeated here.
Step 513:Resource scheduling device is on the external Virtual machine that the virtual machine information received is pointed to, according to the process of the default process migration policy deployment application.
The relevant information of resource scheduling device storage external Virtual machine may include：The corresponding cluster virtual machine information of reference address, external Virtual machine, Resource Properties of external Virtual machine of external Virtual machine etc.；Wherein the Resource Properties of external Virtual machine can be：Outside cloud resource.
Default process migration strategy is for example：The process of pre-determined threshold will be less than in application with other process interaction frequency, i.e. the first process is disposed the internal virtual machine of the cluster virtual machine of the application from internal cloud platform, moved on external Virtual machine.After on the first process migration to external Virtual machine, internal cloud platform disposes the resource that first process of the cluster virtual machine release of the application takes, available for being supplied to other processes of the application to use, to meet the requirement of the application service quality.Further, since the first process and the frequency of other process interactions are relatively low, therefore, by the first process migration to external Virtual machine, the interaction between too many process will not be increased, thus the influence to the performance and service quality of application is smaller.
Or, presetting process migration strategy can also be such as：All processes currently run in the application are disposed on outside virtual machine, and the partial service load of the application is moved into external Virtual machine from the internal virtual machine of cluster virtual machine.So after deployment, the application can carry out load balancing by internal virtual machine and external Virtual machine, thus ensure the service quality of the application.
Or, all processes currently run in an application in multiple applications that the cluster virtual machine is disposed, such as the first application and the application all move to outside void from the internal virtual machine of cluster virtual machine
Plan machine.So after deployment, the resource that first application of cluster virtual machine release takes, provides service, to ensure the service quality of other application available for the other application disposed for the cluster virtual machine, and the business and its service quality of the first application, it can be provided by external Virtual machine.
The present embodiment can be disposed under the scene of multiple applications in a cluster virtual machine, the real time business load for each application disposed to cluster virtual machine, carry out the dynamic dispatching of resource.During the dynamic dispatching of resource, it can determine that internally cloud platform, still to outside cloud platform scheduling resource, improves the flexibility of scheduling of resource according to the traffic load trend of some application.In addition, when the idling-resource of internal cloud platform can not meet the business increased requirement of the application, resource supplement can be carried out to outside cloud platform application resource, and the procedure deployment of external Virtual machine is carried out using flexible process migration strategy, thus ensure the service quality of multiple applications of cluster virtual machine deployment.
Fig. 6 is the structural representation for the resource scheduling device that the embodiment of the present invention five is provided.Resource scheduling device as shown in Figure 6 includes：Data obtaining module 61, number of resources determining module 62, scheduling of resource application module 63 and application process transferring module 64.
Information obtains ear, and block 61 not can be used for the portfolio for obtaining each process currently run in application, and the application deployment is internally in the cluster virtual machine of cloud platform；
When number of resources determining module 62 meets preparatory condition available for the portfolio for each process currently run in the application, according to the flexible strategy of the process of the application obtained in advance, determine that the cluster virtual machine needs the number of resources extended；
Scheduling of resource application module 63 can be used for it is determined that during to outside cloud platform application scheduling resource, sending virtual machine application to outside cloud platform according to the number of resources of determination and asking；
Application process transferring module 64 can be used for after the virtual machine information that the outside cloud platform is returned is received, and the external Virtual machine pointed to the virtual machine information disposes the process of the application.
Optionally, the flexible strategy of the process includes：Increase the application enters number of passes；The cluster virtual machine needs the number of resources extended to be：Increase the number of resources needed for the process of predetermined number.Or, optionally, the flexible strategy of the process includes：Increase the number of resources for the one or more processes currently run in the application；The cluster virtual machine needs the number of resources extended to be：The increased total number resource for needed for each process currently run in the application.
Application process transferring module can carry out process migration deployment according to default process migration strategy.For example：If a cluster virtual machine is deployed with one or more applications, application process transferring module 64 is particularly used in topology relationship between the process according to the application and determines the first process, and by first process from
The internal virtual machine of the cluster virtual machine moves to the external Virtual machine；First process is processes currently being run in the application, with other process interaction frequency less than default frequency.
Or,
If a cluster virtual machine is deployed with one or more applications, application process transferring module 64 is particularly used in each process disposed and currently run in the application on the external Virtual machine, and the partial service load of the application is moved into the external Virtual machine from the internal virtual machine of the cluster virtual machine.
If a cluster virtual machine is deployed with multiple applications, application process transferring module 64 is particularly used in each process that will currently be run in the first application in multiple applications and first application, and the external Virtual machine is moved to from the internal virtual machine of the cluster virtual machine.
Further, resource scheduling device may also include：Scheduling decision module 65.
Scheduling decision module 65 can be used for when the idling-resource of the internal cloud platform can not provide the scheduling of the number of resources of determination, it is determined that to outside cloud platform application scheduling resource.Or, scheduling decision module 65 can be used for the variation tendency for analyzing the portfolio for each process currently run in the application, and be determined according to analysis result to outside cloud platform application scheduling resource.
The present embodiment resource scheduling device can carry out the dynamic dispatching of resource by granularity of process according to the real time business load of the application disposed to cluster virtual machine.During the dynamic dispatching of resource, it can determine that internally cloud platform, still to outside cloud platform scheduling resource, improves the flexibility of scheduling of resource according to the traffic load trend of application.In addition, when the idling-resource of internal cloud platform can not meet the business increased requirement of the application, resource supplement can be carried out to outside cloud platform application resource, and the procedure deployment of external Virtual machine is carried out using flexible process migration strategy, the service quality of application has thus been ensured.The working mechanism of the present embodiment resource scheduling device, reference can be made to the corresponding record of Fig. 1, Fig. 4-5 correspondence embodiment；The structure of resource scheduling device resource scheduling system under mixing cloud mode, reference can be made to the record of Fig. 2-3 correspondence embodiments；It will not be repeated here.
One of ordinary skill in the art will appreciate that：Accompanying drawing is necessary to module or flow in the schematic diagram of one embodiment, accompanying drawing not necessarily implements the present invention.
One of ordinary skill in the art will appreciate that：The module in device in embodiment can be distributed in the device of embodiment according to embodiment description, can also be carried out respective change and be disposed other than in one or more devices of the present embodiment.The module of above-described embodiment can be merged into a module, can also be further split into multiple submodule.
One of ordinary skill in the art will appreciate that：Realize all or part of step of above method embodiment
It can be completed by the related hardware of programmed instruction, foregoing program can be stored in a computer read/write memory medium, the program upon execution, performs the step of including above method embodiment；And foregoing storage medium includes：ROM, RAM, magnetic disc or CD etc. are various can be with the medium of store program codes.
Finally it should be noted that：The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although the present invention is described in detail with reference to the foregoing embodiments, it will be understood by those within the art that：It can still modify to the technical scheme described in previous embodiment, or carry out equivalent substitution to which part technical characteristic；And these modifications or replacement, the essence of appropriate technical solution is departed from the spirit and scope of technical scheme of the embodiment of the present invention.
Claims (1)
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2011/083472 WO2013082742A1 (en) | 2011-12-05 | 2011-12-05 | Resource scheduling method, device and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103503404A true CN103503404A (en) | 2014-01-08 |
Family
ID=48573476
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201180003668.6A Pending CN103503404A (en) | 2011-12-05 | 2011-12-05 | Resource scheduling method, device and system |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN103503404A (en) |
| WO (1) | WO2013082742A1 (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016091035A1 (en) * | 2014-12-08 | 2016-06-16 | 华为技术有限公司 | Resource management method, host, and endpoint |
| CN108108204A (en) * | 2016-11-23 | 2018-06-01 | 湖北省楚天云有限公司 | The application program collocation method and device of cloud computing platform |
| CN110730205A (en) * | 2019-09-06 | 2020-01-24 | 深圳平安通信科技有限公司 | Cluster system deployment method and device, computer equipment and storage medium |
| CN111008064A (en) * | 2019-11-29 | 2020-04-14 | 北京首都在线科技股份有限公司 | Virtual machine resource scheduling method and device, and cluster deployment method and device |
| CN112685179A (en) * | 2020-12-28 | 2021-04-20 | 跬云(上海)信息科技有限公司 | Resource deployment system and method based on cost on cloud |
| CN113760528A (en) * | 2020-12-24 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Resource processing method and device based on multi-cloud platform |
| CN114301987A (en) * | 2022-03-07 | 2022-04-08 | 天津市城市规划设计研究总院有限公司 | Dynamic scheduling method and system for virtualized network resources |
| CN114579295A (en) * | 2020-12-02 | 2022-06-03 | 宏碁股份有限公司 | Method for adaptively calling external hardware resource |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9679253B2 (en) | 2014-11-06 | 2017-06-13 | Copperleaf Technologies Inc. | Methods for maintaining infrastructure equipment and related apparatus |
| CN114363414A (en) * | 2020-09-29 | 2022-04-15 | 华为云计算技术有限公司 | A method, device and system for scheduling computing instances |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100169477A1 (en) * | 2008-12-31 | 2010-07-01 | Sap Ag | Systems and methods for dynamically provisioning cloud computing resources |
| CN101894050A (en) * | 2010-07-28 | 2010-11-24 | 山东中创软件工程股份有限公司 | Method, device and system for flexibly scheduling JEE application resources of cloud resource pool |
| CN101938416A (en) * | 2010-09-01 | 2011-01-05 | 华南理工大学 | A cloud computing resource scheduling method based on dynamic reconfiguration of virtual resources |
| CN102263823A (en) * | 2011-07-25 | 2011-11-30 | 中兴通讯股份有限公司 | Communication method and communication device based on cloud computing |
-
2011
- 2011-12-05 WO PCT/CN2011/083472 patent/WO2013082742A1/en active Application Filing
- 2011-12-05 CN CN201180003668.6A patent/CN103503404A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100169477A1 (en) * | 2008-12-31 | 2010-07-01 | Sap Ag | Systems and methods for dynamically provisioning cloud computing resources |
| CN101894050A (en) * | 2010-07-28 | 2010-11-24 | 山东中创软件工程股份有限公司 | Method, device and system for flexibly scheduling JEE application resources of cloud resource pool |
| CN101938416A (en) * | 2010-09-01 | 2011-01-05 | 华南理工大学 | A cloud computing resource scheduling method based on dynamic reconfiguration of virtual resources |
| CN102263823A (en) * | 2011-07-25 | 2011-11-30 | 中兴通讯股份有限公司 | Communication method and communication device based on cloud computing |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10489207B2 (en) | 2014-12-08 | 2019-11-26 | Huawei Technologies Co., Ltd. | System for resource management using performance specification and description information of the process |
| US11379265B2 (en) | 2014-12-08 | 2022-07-05 | Huawei Technologies Co., Ltd. | Resource management method, host, and endpoint based on performance specification |
| KR20170046786A (en) * | 2014-12-08 | 2017-05-02 | 후아웨이 테크놀러지 컴퍼니 리미티드 | Resource management method, host, and endpoint |
| CN105743808B (en) * | 2014-12-08 | 2017-09-19 | 华为技术有限公司 | A method and device for adapting QoS |
| CN107645407A (en) * | 2014-12-08 | 2018-01-30 | 华为技术有限公司 | A kind of adaptation QoS method and apparatus |
| KR101880407B1 (en) * | 2014-12-08 | 2018-07-19 | 후아웨이 테크놀러지 컴퍼니 리미티드 | Resource management method, host, and endpoint |
| CN105743808A (en) * | 2014-12-08 | 2016-07-06 | 华为技术有限公司 | A method and device for adapting QoS |
| WO2016091035A1 (en) * | 2014-12-08 | 2016-06-16 | 华为技术有限公司 | Resource management method, host, and endpoint |
| US20170199767A1 (en) | 2014-12-08 | 2017-07-13 | Huawei Technologies Co., Ltd. | Resource management method, host, and endpoint |
| CN108108204A (en) * | 2016-11-23 | 2018-06-01 | 湖北省楚天云有限公司 | The application program collocation method and device of cloud computing platform |
| CN110730205A (en) * | 2019-09-06 | 2020-01-24 | 深圳平安通信科技有限公司 | Cluster system deployment method and device, computer equipment and storage medium |
| CN110730205B (en) * | 2019-09-06 | 2023-06-20 | 深圳平安通信科技有限公司 | Cluster system deployment method, device, computer equipment and storage medium |
| CN111008064A (en) * | 2019-11-29 | 2020-04-14 | 北京首都在线科技股份有限公司 | Virtual machine resource scheduling method and device, and cluster deployment method and device |
| CN111008064B (en) * | 2019-11-29 | 2021-10-29 | 北京首都在线科技股份有限公司 | Virtual machine resource scheduling method and device, and cluster deployment method and device |
| CN114579295A (en) * | 2020-12-02 | 2022-06-03 | 宏碁股份有限公司 | Method for adaptively calling external hardware resource |
| CN113760528A (en) * | 2020-12-24 | 2021-12-07 | 北京沃东天骏信息技术有限公司 | Resource processing method and device based on multi-cloud platform |
| CN112685179A (en) * | 2020-12-28 | 2021-04-20 | 跬云(上海)信息科技有限公司 | Resource deployment system and method based on cost on cloud |
| CN114301987A (en) * | 2022-03-07 | 2022-04-08 | 天津市城市规划设计研究总院有限公司 | Dynamic scheduling method and system for virtualized network resources |
| CN114301987B (en) * | 2022-03-07 | 2022-05-20 | 天津市城市规划设计研究总院有限公司 | Dynamic scheduling method and system for virtualized network resources |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2013082742A1 (en) | 2013-06-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103503404A (en) | Resource scheduling method, device and system | |
| US11508021B2 (en) | Processes and systems that determine sustainability of a virtual infrastructure of a distributed computing system | |
| US9705970B2 (en) | System of geographic migration of workloads between private and public clouds | |
| US9558005B2 (en) | Reliable and deterministic live migration of virtual machines | |
| US8638674B2 (en) | System and method for cloud computing | |
| CN105812171B (en) | A virtualized network function VNF control method and device | |
| EP3270289A2 (en) | Container-based multi-tenant computing infrastructure | |
| US20200226520A1 (en) | Methods and systems to optimize server utilization for a virtual data center | |
| US20170109212A1 (en) | Methods and systems to determine and improve cost efficiency of virtual machines | |
| US20130019015A1 (en) | Application Resource Manager over a Cloud | |
| CN104461744A (en) | Resource allocation method and device | |
| CN104917815A (en) | Heterogeneous cloud isolation system and method for in-cloud GIS service computing | |
| US20060031813A1 (en) | On demand data center service end-to-end service provisioning and management | |
| US8959367B2 (en) | Energy based resource allocation across virtualized machines and data centers | |
| CN104011685A (en) | Resource management method of virtual machine system, virtual machine system, and apparatus | |
| CN106201661A (en) | Method and apparatus for elastic telescopic cluster virtual machine | |
| US10481921B2 (en) | Cloud platform, application running method, and access network unit | |
| CN111190691A (en) | Automatic migration method, system, device and storage medium suitable for virtual machine | |
| CN107968810A (en) | A kind of resource regulating method of server cluster, device and system | |
| CN109960579B (en) | Method and device for adjusting service container | |
| CN106664259A (en) | Virtual network function capacity expansion method and apparatus | |
| Zhang et al. | Design and implementation of cloud-based performance testing system for web services | |
| US20180374110A1 (en) | Cost-driven approach to determine inventory layout in cloud computing environments | |
| CN109995571B (en) | Method and device for matching server configuration and VNF application | |
| CN116881012A (en) | Container application vertical capacity expansion method, device, equipment and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140108 |
|
| RJ01 | Rejection of invention patent application after publication |