CN103414767A - Method and device for deploying application software on cloud computing platform - Google Patents
Method and device for deploying application software on cloud computing platform Download PDFInfo
- Publication number
- CN103414767A CN103414767A CN2013103266568A CN201310326656A CN103414767A CN 103414767 A CN103414767 A CN 103414767A CN 2013103266568 A CN2013103266568 A CN 2013103266568A CN 201310326656 A CN201310326656 A CN 201310326656A CN 103414767 A CN103414767 A CN 103414767A
- Authority
- CN
- China
- Prior art keywords
- assemblies
- cloud computing
- computing platform
- node
- assembly
- 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
Images
Landscapes
- Information Transfer Between Computers (AREA)
Abstract
The invention belongs to the field of cloud computing, and discloses a method and device for deploying application systems on a cloud computing platform. The method comprises the steps of configuring the application system into a plurality of assemblies so as to form an assembly array containing the assemblies; computing communication volume between various assemblies in the assembly array; configuring the two assemblies with largest communication volume in the assembly array on a node of the same cluster. According to the method, placing positions of the assemblies on the cloud computing platform are decided according to the communication load between the assemblies of the application system, assemblies with close communication relationship are made to be deployed in the same cluster and the same node meeting the requirements of virtual resources, so that the communication volume inside the cloud computing platform at the utilization time of the application system is reduced.
Description
Technical field
The invention belongs to cloud computing and cloud engineering field, specifically, relate to a kind of method and apparatus that complicated applications Account Dept is deployed in to cloud computing platform.
Background technology
Cloud computing is the pattern of emerging in recent years a kind of interactive service, with as required, the mode of easily expansion provides the required resource of user or service.After the concept of cloud computing proposed, academia and industrial quarters had all been launched research widely, and each large IT vendor has also released the cloud computing product of oneself successively.
The cloud computing in the whole world be take the U.S. as main, and the IT companies such as Google, Amazon, IBM, Microsoft have all issued cloud computing product separately.The search engine of Google just is based upon on the support that is distributed in 1,000,000 station servers all over the world, and these infrastructure have also been used in the Google earth, map, mail service etc.Amazon, immediately following the paces of Google, is used elasticity calculating cloud (EC2) and simple storage service (S3) to provide and calculate and stores service as the enterprises and individuals.After this, the Windows Azure cloud operating system of " Lan Yun " cloud computing platform of IBM, Microsoft etc. has all completed research and development and has promoted the use of.
In existing cloud computing product, Microsoft provides Windows Azure cloud operating system for application system migration and deployment, Amazon EC2 be take the mode of publicly-owned cloud IAAS provides virtual machine to be user's application deployment, Google GAE and the SAE of the Sina more deployment of application system provide the environment of developing and using, and proposition and the application of the concept such as " mixed cloud " " virtual application device ", " layout of cloud service ability " and technology all are being deployed to this problem of cloud computing platform round application system.
The application system pastern is deployed on cloud computing platform, can adopt traditional deployment way based on virtual machine, as on Amazon EC2 platform, by the deployment personnel to EC2 platform application virtual machine, after by network, be connected to corresponding virtual machine, carry out deployment and the configuration of application system database, WEB container and code etc.In addition, the high cloud computing platform of product provides the deployment way based on the application system deployment scheme, the deployment personnel create the deployment scheme of application system by product high cloud platform, and upload code and the data of each assembly of application system, by product high cloud platform backstage, realize the establishment of virtual machine and the deployment of application, after application system and virtual machine thereof that deployment is completed return to the deployment personnel, personnel further configure and test by deployment.
The complicated applications system adopts Componentized development deployment mode usually at present, application software can be deployed in modules take Intel Virtualization Technology in the virtual machine of the cloud computing platform of core, and will be deployed in a virtual machine, the system module with specific function is defined as an assembly of system.Application system for complicated multicompartment, at deployment of components in the process of cloud platform, a key issue that needs to solve is reasonably to select target cloud virtual machine (Cloud Virtual Machine, CVM) cluster and the node thereof on the cloud platform for each assembly.
To this problem, the researcher has also made many fruitful effort, and pertinent literature discloses as follows:
In Chinese patent 102932418A, disclose a kind of cloud application deployment method, the method is by obtaining the load of current virtual machine, and the service request load of each functional module of its medium cloud application; Higher limit and the lower limit of virtual machine load are set, and the higher limit of service request load and lower limit; In the situation that current virtual machine load is higher than virtual machine load higher limit, a newly-built target virtual machine, be deployed to the functional module on current virtual machine in described target virtual machine.In the situation that current virtual machine load is lower than virtual machine load lower limit, according to the higher limit of service request load and the service request load of each functional module, judge the service request load that whether there is at least one functional module on the current virtual machine higher limit higher than the service request load; If not, other virtual machines of this current virtual machine and satisfied merging condition are merged.
Chinese patent 102281329A discloses the resource regulating method that a kind of platform is namely served (PAAS) cloud computing platform, in cloud computing platform, management node detects the loading condition of each child node, the child node that surpasses threshold value for load, application by load expense maximum in described child node, be re-deployed to load the gentliest and the child node of not disposing described application.
Usually, cloud computing platform decides the placement location of assembly or virtual machine by resource dispatching strategy, as adopt greedy strategy, can be by assembly or the deploying virtual machine node that surplus resources is maximum in the maximum CVM cluster of surplus resources and cluster, adopt to save strategy, can be by assembly or the minimum node of deploying virtual machine surplus resources in the minimum CVM cluster of surplus resources and cluster.Yet resource dispatching strategy not take an application system and is unit, does not consider the correspondence of each inter-module of application system yet.If the multicompartment application system is not considered the relation of inter-module in deployment, can cause when deployment system, make each assembly under specific resource dispatching strategy, be deployed to respectively on the different nodes of different segment, different CVM clusters in the cloud platform, make application system in use, cause the mass communication of the inner cross-network segment of cloud platform, i.e. the waste of cloud platform internal bandwidth.
Summary of the invention
For the problems referred to above, the objective of the invention is to provide a kind of method and apparatus that application software is deployed in to cloud computing platform, the method has been considered the correspondence between the assembly of application system, by correspondence closely assembly be deployed on the node of same cluster of cloud platform as far as possible, with the quick response that realizes application system, the demand that reduces cloud platform traffic load.
Concrete, the technical solution used in the present invention is as follows:
First aspect present invention provides a kind of application system has been deployed in to the method on cloud computing platform, comprising:
Application system is configured to several assemblies, forms the column of assemblies that comprises described assembly;
Calculate the traffic between each assembly in described column of assemblies;
By on the node of two arrangement of components in same cluster of traffic maximum in described column of assemblies.
As a kind of improvement of technical solution of the present invention, the method also comprises:
Two assemblies having disposed are merged into to an assembly, and in described column of assemblies in other assembly form new column of assemblies;
By on the node of two arrangement of components in same cluster of traffic maximum in new column of assemblies.
According to the method that first aspect present invention provides, using two assemblies before merging and other component communication amounts and as the traffic of described new assembly.
According to the method that first aspect present invention provides, the method also comprises:
If on described cloud computing platform, there is the node of the resource requirement sum of two assemblies that meet traffic maximum, described two assemblies are configured on this node.
The method provided according to first aspect present invention, if on described cloud computing platform, there is the node of the resource requirement sum of two or more two assemblies that meet traffic maximum, according to the node of described two assemblies of resource dispatching strategy option and installment of described cloud computing platform.
According to the method that first aspect present invention provides, the method also comprises:
Calculate the resource requirement value of described application system, and select available total resources in described cloud computing platform to be greater than the cluster of the resource requirement value of described application system;
According to the resource dispatching strategy of described cloud computing platform, select cluster.
Corresponding, second aspect present invention also provides a kind of application software has been deployed in to the device on cloud computing platform, and this device comprises:
Column of assemblies forms module, for application software configuration being become to several assemblies, forms the column of assemblies that comprises described assembly;
The traffic volume calculations module, for calculating the traffic between each assembly of described column of assemblies;
The arrangement of components module, for the node of two arrangement of components in same cluster by described column of assemblies traffic maximum.
Device according to second aspect present invention provides also comprises:
New Parent row form module, merge into an assembly for two assemblies by having disposed, and in described column of assemblies in other assembly form new column of assemblies;
The New Parent configuration module, by the node of two arrangement of components in same cluster of traffic maximum in new column of assemblies.
According to the device that second aspect present invention provides, using two assemblies before merging and other component communication amounts and as the traffic of described new assembly.
Device according to second aspect present invention provides also comprises:
The Resource Calculation module, be used to calculating the resource requirement value of described application system, and select available total resources in described cloud computing platform to be greater than the cluster of the resource requirement value of described application system;
Cluster is selected module, selects cluster for the resource dispatching strategy according to described cloud computing platform.
With respect to prior art, the present invention by correspondence closely assembly be deployed on the node of same cluster of cloud platform as far as possible, with the quick response that realizes application system, the demand that reduces cloud platform traffic load.
In addition, the present invention by correspondence closely deployment of components on the node of the same cluster of cloud platform in, can also dispose according to the resource dispatching strategy that cloud computing platform is used, meet the resource dispatching strategy of cloud computing platform.
The accompanying drawing explanation
Fig. 1 is based on the configuration diagram of the cloud computing platform of Eucalyptus, in figure, can find out, cloud computing platform adopts the framework of layering, three clusters of a cloud controller (CLC) management, three clusters have respectively three, four and two nodes, the application system deployment module realizes method of the present invention, and, by calling the interface of CLC, realizes each deployment of components of application system on the node of Node Controller (NC) management;
Fig. 2 is the flow chart that application system is deployed in to an embodiment of the method on cloud computing platform of the present invention;
Fig. 3 is the flow chart that application system is deployed in to an embodiment of the method on cloud computing platform of the present invention;
Fig. 4 is the flow chart that application system is deployed in to an embodiment of the method on cloud computing platform of the present invention;
Fig. 5 is the flow chart that application system is deployed in to an embodiment of the method on cloud computing platform of the present invention;
Fig. 6 is the structural representation that application system is deployed in to an embodiment of the device on cloud computing platform of the present invention;
Fig. 7 is the structural representation that application system is deployed in to an embodiment of the device on cloud computing platform of the present invention;
Fig. 8 is the structural representation that application system is deployed in to an embodiment of the device on cloud computing platform of the present invention;
In figure:
10: the application system deployment module; 20: cloud controller; 30: cluster controller; 40: Node Controller; 601: column of assemblies forms module; 602: the traffic volume calculations module; 603: the arrangement of components module; 604: the New Parent row form module; 605: the New Parent configuration module; 606: the Resource Calculation module; 607: cluster is selected module.
Embodiment
Below by object lesson, further illustrate the present invention.But, should be understood to, example is only used for the use specifically described more in detail, and should not be construed as for limiting in any form the present invention.
Although following embodiments of the invention are based on the high cloud computing platform of the product of Eucalyptus framework, but be to be understood that, embodiments of the invention also can move on other cloud computing platforms, such as AmazonEC2, Huawei's cloud computing platform and the privately owned cloud of EMC etc.
Referring to Fig. 1, Fig. 1 is based on the configuration diagram of the cloud computing platform of Eucalyptus.In the execution mode shown in Fig. 1, the high cloud computing platform of the product of Eucalyptus framework is used product high cloud version 3 .2.4.Cloud computing platform has adopted the architecture of layering, wherein, represents cloud controller 20(Cloud Controller, CLC) be in the first floor, the second layer is cluster controller 30(Cluster Controller, CC), the 3rd layer is Node Controller 40(Node Controller, NC).Cloud controller 20 is responsible for accepting virtual resource management request, and by the management to cluster controller 30, realizes the scheduling of virtual resource, and wherein, a cloud controller 20 can be dispatched the resource of a plurality of cluster controllers 30.A cluster controller 30 can be managed a plurality of Node Controllers 40, by the resource state information of Node Controller 40 collector nodes, and virtual resource is processed to request and carry out to each Node Controller 40; Node Controller 40 is in charge of a node, and node is the physical server through virtualization process.Node Controller 40 is by receiving the order of cluster controller 30, and execution starts, checks, closes and remove the virtual resource of node etc.Although in the cloud computing platform shown in Fig. 1, a cloud controller is connected with three cluster controllers 30 for 20 times, three cluster controllers are connected with respectively three, four and two Node Controllers 40 for 30 times, but be to be understood that, cluster controller 30 in cloud computing platform and Node Controller 40 can be a plurality of arbitrarily, those skilled in the art can arrange according to actual needs, and the present invention is not restricted this.
As shown in Figure 1, some preferred embodiment in the middle of, need in the cloud platform architecture, add application system deployment module 10, in order to receive application system, dispose request, and, by calling the interface of cloud controller 20, realize each deployment of components of application system on the node of Node Controller 40 management.Generally speaking, the available virtual resource of cloud computing platform should be much larger than the virtual resource of application system demand, and namely application system can find the CVM cluster that meets resource requirement on the cloud platform.
Referring to Fig. 2, Fig. 2 is the flow chart that application system is deployed in to an embodiment of the method on cloud computing platform of the present invention.In the execution mode shown in Fig. 2, the method comprises the steps:
Step S201: application system is configured to several assemblies, forms the column of assemblies that comprises described assembly.
Step S202: calculate the traffic between each assembly in described column of assemblies, the traffic be specially inter-module per second transmitted in both directions byte number and, unit is byte number/second; The component communication amount is calculated with the Actual metering on kinetic state method, completes in running environment that can be before migration is disposed or test environment.
According to the correspondence of inter-module, press inter-component communication amount size, assembly, to sequence, is set up to the right ordered sequence of assembly.Being provided with the order sequence is S,
S={<C
i, C
j>/C
i, C
j∈ V} ∩ element is pressed the sequence of inter-component communication amount size;
Wherein, C proxy component, V represent the set of application system assembly.
In some embodiments, suppose that certain application software comprises four assembly C1～C4, sort according to the size of the correspondence between C1～C4 and the traffic, form component sequence S, wherein:
S={<C1,C2>,<C2,C4>,<C1,C4>,<C1,C3>,<C3,C4>}。
Step S203: by the node of two arrangement of components in same cluster of traffic maximum in described column of assemblies, on the same node of preferred disposition in same cluster.For example, in the above-described embodiment, C1 and C2 can be deployed on the same node in same cluster.
Apparent, the method that application software is deployed in to cloud computing platform in this execution mode has been considered the correspondence between the assembly of application system, by correspondence closely assembly be deployed on the node of same cluster of cloud platform as far as possible, thereby can realize application system quick response, reduce the demand of cloud platform traffic load.
In execution mode shown in Fig. 2 by the node of two arrangement of components in same cluster of the traffic maximum in component sequence.For other assemblies in component sequence, those skilled in the art can adopt the means of knowing to be configured, for example can be according to the load of the current virtual machine obtained, and wherein the resource requirement of each assembly of cloud computing platform is configured, also can detect according to management node in cloud computing platform the loading condition of each child node, assembly by load expense maximum in described child node, be re-deployed to load the gentliest and the child node of not disposing described application.
Yet, some preferred embodiment in, can also other assemblies in component sequence be configured again according to the correspondence of assembly and the size of the traffic, further improve the demand that assembly is realized the quick response of application system, reduced cloud platform traffic load.
Referring to Fig. 3, Fig. 3 is the flow chart that application system is deployed in to an embodiment of the method on cloud computing platform of the present invention.In the method shown in Fig. 3, the step S301 of the method～S303 correspondence is identical.In addition, the method also comprises step S304 and S305, is configured for the assembly that component sequence is not yet configured.
Step S304: two assemblies having disposed are merged into to an assembly, and in described column of assemblies in other assembly form new column of assemblies.
In some preferred implementation, can by the assembly after merging and other component communication amounts for merge front two assemblies and other component communication amounts with, then according to inter-component communication amount size, to assembly to sorting.For example, in above-mentioned example, by after assembly C1 and C2 deployment, assembly C1 and C2 merging are become to a new assembly C12, the size of its traffic is C1 and other component communication amounts and C2 and other component communication amount sums, for example, the traffic between C12 and C4 be updated to former C1 and C2 respectively with the traffic of C4 and, successively new component sequence (C12, C3, C4) is sorted, the new component sequence S after renewal is:
S={<C12,C4><C12,C3><C3,C4>}。
Step S305: by the node of two arrangement of components in same cluster of traffic maximum in new column of assemblies.For example, in above-mentioned example, can accordingly C12 and C4 be deployed on the same node in same cluster.Apparent, can also further improve the communication efficiency between assembly like this, improved the response speed of application program.
Some preferred embodiment in, the method can also repeat above-mentioned several steps traversal ordered sequences and dispose.After each deployment of components, again travel through the ordered sequence S after upgrading, until the assembly in ordered sequence S is not to meeting the deployment conditions of calculating.Generally speaking, if the assembly of assembly centering contains and has disposed, by this deployment of components to this node, and upgrade ordered sequence S, the method for renewal is the same.For example, in above-mentioned example, the assembly of C12 for having disposed, if C12 place node meets the resource requirement of C4, be deployed to C4 the virtual machine that this node meets resource requirement, the ordered sequence S={<C124 after renewal, C3 > }.If the assembly of assembly centering is all not dispose, can dispose according to other requirements of assembly and cluster, for example can calculate in cluster whether have the node surplus resources meet two assembly CPU, internal memory, hard disk, bandwidth resources requirements and, if there is the node that meets the node resource demand, two assemblies are deployed to respectively in two virtual machines that this node meets resource, and upgrade ordered sequence S.For example in above-mentioned example, have ordered sequence S, wherein,
S={<C2,C4>,<C1,C4>,<C1,C3>,<C3,C4>}；
If the assembly in ordered sequence S is right<C2, C4 >,<C1, C4,<C1, C3 > do not meet the deployment conditions of calculating, and cluster has node to meet C3, the C4 resource requirement and, by C3, C4 is deployed to respectively in two virtual machines of this node.Complete the assembly in traversal ordered sequence S, if application system still has assembly not dispose, each assembly can be deployed in separately in cluster on the node that meets resource requirement, if the quantity of node is greater than 1, according to cloud platform resource scheduling strategy, select optimal node.
The result that traversal ordered sequence S disposes in cluster is that as far as possible that the traffic is large assembly is to being deployed on same node.
In addition, the method that provides of the embodiment of the present invention can also realize automatically the configuration to component resources demand and node resource.For example by deployment of components before node, can also calculate the surplus resources of node and the resource requirement of assembly, if on described cloud computing platform, there is the node of the resource requirement sum of two assemblies that meet traffic maximum, described two assemblies are configured on this node.
However, in most cases, due to the resource that can provide on cloud computing platform, such as CPU, internal memory, hard disk, bandwidth etc. is much larger than the resource requirement of application software, therefore, generally can in cluster, find two, or a plurality of surplus resources node meets the resource requirement of assembly.Referring to Fig. 4, Fig. 4 is the flow chart that application system is deployed in to an embodiment of the method on cloud computing platform of the present invention.In the execution mode shown in Fig. 4, in step S401～S404 and the execution mode shown in Fig. 3 step S301～S304 is corresponding identical.In addition, in the execution mode shown in Fig. 4, also comprise step S405.
Wherein, step S405: if there is the node of the resource requirement sum of two or more two assemblies that meet traffic maximum on described cloud computing platform, according to the node of described two assemblies of resource dispatching strategy option and installment of described cloud computing platform.To this, be discussed in detail below.
Computing application system and each assembly are to CPU, MEM(internal memory), the HD(hard disk) and the BW(bandwidth) the resource requirement value, wherein the bandwidth demand value comprises two parts, BW=BW1+BW2, the requirements of BW1 for communicating by letter with the application system intraware, the requirements of BW2 for communicating by letter with the external system user, two parts value sum is total bandwidth demand value.If in cluster, have node meet two component resources requirements and, establish in cluster the resource requirement value that meets assembly and the set of node be N ':
If | N ' |=1, there is a node to meet, continue step S406, two assemblies are moved to respectively to this node and meet in two virtual machines of resource, upgrade ordered sequence S.Wherein step S406 is identical with the step S305 in the execution mode shown in Fig. 3, no longer repeats herein.
If | N ' | > 1, there are a plurality of nodes to meet, need from most suitable node of the inner selection of N ', the resource dispatching strategy that the method for selection need be current according to cloud computing platform, from N ', select resource maximum, minimum or other meet the node of cloud platform resource scheduling.Then, continue step S406, two assemblies are deployed to respectively in two virtual machines that this node meets resource, upgrade ordered sequence S.Wherein step S406 is identical with the step S305 in the execution mode shown in Fig. 3, no longer repeats herein.
In some cases, if traveled through the assembly pair of S, | N ' | be all 0, select successively the assembly pair in ordered sequence S, until in cluster, exist two nodes to meet respectively the virtual resource demand of assembly, these two assemblies are deployed to respectively in the virtual machine on two nodes that meet resource requirement, upgrade ordered sequence S.For example in above-mentioned example, ordered sequence S is:
S={<C1,C2>,<C2,C4>,<C1,C4>,<C1,C3>,<C3,C4>}；
Suppose in cluster not have node meet the right resource requirement of all component in the S sequence and, but two nodes that meet respectively C1 and C2 resource requirement are arranged, dispose respectively C1 and C2, the ordered sequence S={<C2 after renewal, C4 >,<C1, C4 >,<C1, C3 >,<C3, C4 >.
In addition, in some preferred implementation, can also, at the deployment of components by application software before node, also comprise the step of selecting the cluster in cloud computing platform.Referring to Fig. 5, Fig. 5 is the flow chart that application system is deployed in to an embodiment of the device on cloud computing platform of the present invention.In the execution mode shown in Fig. 5, the method also comprises step S501～S507.
Step S501: calculate the resource requirement value of described application system, and be greater than in the cluster of resource requirement value of described application system from available total resources described cloud computing platform, select cluster according to the resource dispatching strategy of described cloud computing platform.Below choice set group's step is discussed in detail.
Some preferred embodiment in, can at first calculate in cloud computing platform, the available resource value of each CVM cluster, for example include but not limited to total surplus amount and the average surplus of CPU, internal memory, hard disk, bandwidth, that is:
Wherein, Cloud is f the cluster that cloud computing platform comprises, and NodeCCi is the nodes of i CVM cluster, and SumCPUCCi is the CPU sum of i CVM cluster, AvgCPUCCi is the CPU average of i CVM cluster, and internal memory, hard disk, bandwidth are with the calculating of CPU.
Then, the CVM cluster is selected, realized from a plurality of CVM clusters of cloud computing platform, select the target that a cluster is application system to dispose cluster.From cloud computing platform, calculating the cluster that the surplus resources total amount is greater than application system resource requirement value in all CVM clusters, set up cluster set F ', calculate the number of cluster set F ' element.
If cluster set F ' quantity is 1, select the cluster in set.
If cluster set F ' quantity is greater than 1, need in the cluster set to select a most suitable cluster, the method of selecting can be the resource dispatching strategy current according to the cloud platform, as adopt load balancing, can select the maximum cluster of average resource total amount in the cluster set, as adopt the strategy that resource utilization is the highest, can select the minimum cluster of average resource total amount in the cluster set.
In addition, the step S401 in step S502～507 and the execution mode shown in Fig. 4～406 are corresponding identical, no longer repeat herein.
Corresponding, the embodiment of the present invention also provides a kind of application software has been deployed in to the device on cloud computing platform.Referring to Fig. 6, Fig. 6 is the structural representation that application system is deployed in to an embodiment of the device on cloud computing platform of the present invention.In the execution mode shown in Fig. 6, this device comprises:
Column of assemblies forms module 601, for application software configuration being become to several assemblies, forms the column of assemblies that comprises described assembly;
Traffic volume calculations module 602, for calculating the traffic between each assembly of described column of assemblies;
Arrangement of components module 603, for the node of two arrangement of components in same cluster by described column of assemblies traffic maximum.
Wherein, traffic volume calculations module 602, according to the traffic size between the assembly in column of assemblies, to sequence, is set up the right ordered sequence of assembly to assembly.Being provided with the order sequence is S,
S={<C
i, C
j>/C
i, C
j∈ V} ∩ element is pressed the sequence of inter-component communication amount size;
Wherein, C proxy component, V represent the set of application system assembly.
In some embodiments, suppose that certain application software comprises four assembly C1～C4, sort according to the size of the correspondence between C1～C4 and the traffic, form component sequence S, wherein:
S={<C1,C2>,<C2,C4>,<C1,C4>,<C1,C3>,<C3,C4>}。
After completing sequence, arrangement of components module 603 is by the node of two arrangement of components in same cluster of traffic maximum in described column of assemblies.For example, in the above-described embodiment, C1 and C2 can be deployed on the same node in same cluster.
Apparent, the device that application software is deployed in to cloud computing platform in this execution mode has been considered the correspondence between the assembly of application system, by correspondence closely assembly be deployed on the node of same cluster of cloud platform as far as possible, thereby can realize application system quick response, reduce the demand of cloud platform traffic load.
Yet, some preferred embodiment in, can also other assemblies in component sequence be configured again according to the correspondence of assembly and the size of the traffic, further improve the demand that assembly is realized the quick response of application system, reduced cloud platform traffic load.
Referring to Fig. 7, Fig. 7 is the structural representation that application system is deployed in to an embodiment of the device on cloud computing platform of the present invention.In the execution mode shown in Fig. 7, this device also comprises:
New Parent row form module 604, merge into an assembly for two assemblies by having disposed, and in described column of assemblies in other assembly form new column of assemblies;
New Parent configuration module 605, by the node of two arrangement of components in same cluster of traffic maximum in new column of assemblies.Wherein, in some preferred implementation, using two assemblies before merging and other component communication amounts and as the traffic of described new assembly.
Some preferred embodiment in, the method can also repeat above-mentioned several steps traversal ordered sequences and dispose.After each deployment of components, again travel through the ordered sequence after upgrading, until the assembly in ordered sequence is not to meeting the deployment conditions of calculating.
Referring to Fig. 8, Fig. 8 is the structural representation that application system is deployed in to an embodiment of the device on cloud computing platform of the present invention.In the execution mode shown in Fig. 8, this device also comprises:
Resource Calculation module 606, be used to calculating the resource requirement value of described application system, and select available total resources in described cloud computing platform to be greater than the cluster of the resource requirement value of described application system;
Cluster is selected module 607, selects cluster for the resource dispatching strategy according to described cloud computing platform.
The foregoing is only the preferred embodiments of the present invention; not thereby limit the scope of the claims of the present invention; every equivalent structure or equivalent flow process conversion that utilizes specification of the present invention and accompanying drawing content to do; or directly or indirectly be used in other relevant technical fields, all in like manner be included in scope of patent protection of the present invention.
Claims (10)
1. one kind is deployed in the method on cloud computing platform by application system, it is characterized in that, comprising:
Application system is configured to several assemblies, forms the column of assemblies that comprises described assembly;
Calculate the traffic between each assembly in described column of assemblies;
By on the node of two arrangement of components in same cluster of traffic maximum in described column of assemblies.
2. as claimed in claim 1 application system is deployed in to the method on cloud computing platform, it is characterized in that, also comprise:
Two assemblies having disposed are merged into to an assembly, and in described column of assemblies in other assembly form new column of assemblies;
By on the node of two arrangement of components in same cluster of traffic maximum in new column of assemblies.
3. as claimed in claim 2 application system is deployed in to the method on cloud computing platform, it is characterized in that, using two assemblies before merging and other component communication amounts and as the traffic of described new assembly.
4. as claimed in claim 1 application system is deployed in to the method on cloud computing platform, it is characterized in that, also comprise:
If on described cloud computing platform, there is the node of the resource requirement sum of two assemblies that meet traffic maximum, described two assemblies are configured on this node.
5. as claimed in claim 4 application system is deployed in to the dispositions method on cloud computing platform, it is characterized in that,
If on described cloud computing platform, there is the node of the resource requirement sum of two or more two assemblies that meet traffic maximum, according to the node of described two assemblies of resource dispatching strategy option and installment of described cloud computing platform.
6. as claimed in claim 1 application system is deployed in to the dispositions method on cloud computing platform, it is characterized in that, also comprise:
Calculate the resource requirement value of described application system, and be greater than in the cluster of resource requirement value of described application system from available total resources described cloud computing platform, select cluster according to the resource dispatching strategy of described cloud computing platform.
7. one kind is deployed in the device on cloud computing platform by application software, it is characterized in that, comprising:
Column of assemblies forms module, for application software configuration being become to several assemblies, forms the column of assemblies that comprises described assembly;
The traffic volume calculations module, for calculating the traffic between each assembly of described column of assemblies;
The arrangement of components module, for the node of two arrangement of components in same cluster by described column of assemblies traffic maximum.
8. as claimed in claim 7 application software is deployed in to the device on cloud computing platform, it is characterized in that, also comprise:
New Parent row form module, merge into an assembly for two assemblies by having disposed, and in described column of assemblies in other assembly form new column of assemblies;
The New Parent configuration module, by the node of two arrangement of components in same cluster of traffic maximum in new column of assemblies.
9. as claimed in claim 8 application software is deployed in to the device on cloud computing platform, it is characterized in that,
Using two assemblies before merging and other component communication amounts and as the traffic of described new assembly.
10. as claimed in claim 6 application software is deployed in to the device on cloud computing platform, it is characterized in that, also comprise:
The Resource Calculation module, be used to calculating the resource requirement value of described application system, and select available total resources in described cloud computing platform to be greater than the cluster of the resource requirement value of described application system;
Cluster is selected module, selects cluster for the resource dispatching strategy according to described cloud computing platform.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013103266568A CN103414767A (en) | 2013-07-30 | 2013-07-30 | Method and device for deploying application software on cloud computing platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013103266568A CN103414767A (en) | 2013-07-30 | 2013-07-30 | Method and device for deploying application software on cloud computing platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103414767A true CN103414767A (en) | 2013-11-27 |
Family
ID=49607753
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013103266568A Pending CN103414767A (en) | 2013-07-30 | 2013-07-30 | Method and device for deploying application software on cloud computing platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103414767A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104219285A (en) * | 2014-08-12 | 2014-12-17 | 重庆大学 | Method for determining mapping relation of communication agent nodes and virtual machines in cloud platform |
| CN104967661A (en) * | 2015-05-12 | 2015-10-07 | 华南师范大学 | Assembly deploying method and device used for application system mixing deployment under cloud environment |
| CN105897826A (en) * | 2015-11-24 | 2016-08-24 | 乐视云计算有限公司 | Cloud platform service creating method and system |
| CN106325998A (en) * | 2015-06-30 | 2017-01-11 | 华为技术有限公司 | Method and device for deploying application based on cloud computing |
| CN106537322A (en) * | 2014-06-30 | 2017-03-22 | 微软技术许可有限责任公司 | Effective range partition splitting in scalable storage |
| CN108228272A (en) * | 2016-12-22 | 2018-06-29 | 中国移动通信集团上海有限公司 | WEB containers generation processing method, equipment and server |
| CN109976873A (en) * | 2019-02-25 | 2019-07-05 | 华中科技大学 | The scheduling scheme acquisition methods and dispatching method of containerization distributed computing framework |
| CN110221920A (en) * | 2019-06-04 | 2019-09-10 | 合肥讯飞数码科技有限公司 | Dispositions method, device, storage medium and system |
| CN112579987A (en) * | 2020-12-04 | 2021-03-30 | 河南大学 | Migration deployment method and operation identity verification method of remote sensing program in hybrid cloud |
| CN114924846A (en) * | 2022-07-22 | 2022-08-19 | 浙江云针信息科技有限公司 | Virtual machine migration method based on cloud operating system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7392314B2 (en) * | 2003-08-15 | 2008-06-24 | International Business Machines Corporation | System and method for load—balancing in a resource infrastructure running application programs |
| CN101753359A (en) * | 2009-12-25 | 2010-06-23 | 用友软件股份有限公司 | Method and system for dynamically distributing components |
| CN101860752A (en) * | 2010-05-07 | 2010-10-13 | 浙江大学 | Video code stream parallelization method for embedded multi-core system |
-
2013
- 2013-07-30 CN CN2013103266568A patent/CN103414767A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7392314B2 (en) * | 2003-08-15 | 2008-06-24 | International Business Machines Corporation | System and method for load—balancing in a resource infrastructure running application programs |
| CN101753359A (en) * | 2009-12-25 | 2010-06-23 | 用友软件股份有限公司 | Method and system for dynamically distributing components |
| CN101860752A (en) * | 2010-05-07 | 2010-10-13 | 浙江大学 | Video code stream parallelization method for embedded multi-core system |
Non-Patent Citations (1)
| Title |
|---|
| 张靓: "云平台应用系统迁移方法的研究", 《计算机科学》 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106537322A (en) * | 2014-06-30 | 2017-03-22 | 微软技术许可有限责任公司 | Effective range partition splitting in scalable storage |
| CN106537322B (en) * | 2014-06-30 | 2020-03-13 | 微软技术许可有限责任公司 | Efficient range partition splitting in scalable storage |
| CN104219285B (en) * | 2014-08-12 | 2018-05-29 | 重庆大学 | The method for determining the mapping relations of communication agent node and virtual machine in cloud platform |
| CN104219285A (en) * | 2014-08-12 | 2014-12-17 | 重庆大学 | Method for determining mapping relation of communication agent nodes and virtual machines in cloud platform |
| CN104967661A (en) * | 2015-05-12 | 2015-10-07 | 华南师范大学 | Assembly deploying method and device used for application system mixing deployment under cloud environment |
| CN104967661B (en) * | 2015-05-12 | 2018-04-20 | 华南师范大学 | Deployment of components method and apparatus applied to application system mixed deployment under cloud environment |
| CN106325998B (en) * | 2015-06-30 | 2020-08-25 | 华为技术有限公司 | Application deployment method and device based on cloud computing |
| CN106325998A (en) * | 2015-06-30 | 2017-01-11 | 华为技术有限公司 | Method and device for deploying application based on cloud computing |
| CN105897826A (en) * | 2015-11-24 | 2016-08-24 | 乐视云计算有限公司 | Cloud platform service creating method and system |
| CN108228272A (en) * | 2016-12-22 | 2018-06-29 | 中国移动通信集团上海有限公司 | WEB containers generation processing method, equipment and server |
| CN108228272B (en) * | 2016-12-22 | 2021-06-08 | 中国移动通信集团上海有限公司 | WEB container generation processing method, equipment and server |
| CN109976873A (en) * | 2019-02-25 | 2019-07-05 | 华中科技大学 | The scheduling scheme acquisition methods and dispatching method of containerization distributed computing framework |
| CN109976873B (en) * | 2019-02-25 | 2020-12-18 | 华中科技大学 | Scheduling scheme obtaining method and scheduling method of containerized distributed computing framework |
| CN110221920A (en) * | 2019-06-04 | 2019-09-10 | 合肥讯飞数码科技有限公司 | Dispositions method, device, storage medium and system |
| CN110221920B (en) * | 2019-06-04 | 2022-02-18 | 合肥讯飞数码科技有限公司 | Deployment method, device, storage medium and system |
| CN112579987A (en) * | 2020-12-04 | 2021-03-30 | 河南大学 | Migration deployment method and operation identity verification method of remote sensing program in hybrid cloud |
| CN112579987B (en) * | 2020-12-04 | 2022-09-13 | 河南大学 | Migration deployment method and operation identity verification method of remote sensing program in hybrid cloud |
| CN114924846A (en) * | 2022-07-22 | 2022-08-19 | 浙江云针信息科技有限公司 | Virtual machine migration method based on cloud operating system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103414767A (en) | Method and device for deploying application software on cloud computing platform | |
| CN103109271B (en) | The implementation method of migrate application and system between a kind of platform | |
| CN108737168B (en) | Container-based micro-service architecture application automatic construction method | |
| CN102857370B (en) | A kind of method of Resources allocation and device | |
| CN103870314A (en) | Method and system for simultaneously operating different types of virtual machines by single node | |
| CN112783649A (en) | Cloud computing-oriented interactive perception containerized micro-service resource scheduling method | |
| CN103064742A (en) | Automatic deployment system and method of hadoop cluster | |
| CN102981910A (en) | Realization method and realization device for virtual machine scheduling | |
| CN103873534A (en) | Method and device for application cluster migration | |
| Gogouvitis et al. | Seamless computing in industrial systems using container orchestration | |
| CN105071994B (en) | A kind of mass data monitoring system | |
| CN101937359A (en) | Simulation application-orientated universal extensible computing system | |
| CN100588197C (en) | Gridding emulation method and its emulator | |
| CN115134371A (en) | Scheduling method, system, equipment and medium containing edge network computing resources | |
| CN103677983A (en) | Scheduling method and device of application | |
| CN103023936A (en) | Multi-hierarchy network system and task executing method based on same | |
| CN105488288A (en) | NS3 (Network Simulator Version-3) parallel analog simulation system | |
| CN106155822A (en) | A kind of disposal ability appraisal procedure and device | |
| CN112261125B (en) | Centralized unit cloud deployment method, device and system | |
| CN104991826A (en) | Method and apparatus for deploying virtual machine | |
| CN112328402A (en) | High-efficiency self-adaptive space-based computing platform architecture and implementation method thereof | |
| Abase et al. | Locality sim: cloud simulator with data locality | |
| Zhang et al. | Repeatable multi-dimensional virtual network embedding in cloud service platform | |
| CN110868330A (en) | Evaluation method, device and evaluation system for CPU resources which can be divided by cloud platform | |
| CN112241304B (en) | Loongson cluster super-fusion resource scheduling method and device and Loongson cluster |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20131127 |