JP2012181655A - Module deployment device, deployment method, and deployment program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a module deployment device, deployment method, and deployment program that are capable of deploying a module with an appropriate sharing degree corresponding to requirements or the like.SOLUTION: A module deployment device compares information on a module sequence, sharability information, and information on modules deployed in server devices to retrieve a server device whose sharability information in the highest layer of modules deployed in the server indicates being deployable, and outputs information on a module in a higher layer in the module sequence than the modules deployed in the retrieved server device.

Description

  The present invention relates to a module deployment device, a deployment method, and a deployment program.

  The module deployment system to the server device or the like is based on the premise that specific middleware exists in the deployment destination server device. However, in an environment where various services are operating on a large amount of server resources, middleware that is a prerequisite for the operation of the service may not be uniform. There is also a situation in which a plurality of service applications share modules such as hardware, OS, and MW (Middle Ware) for the purpose of resource aggregation. However, it is difficult to deploy modules in an environment where services that can be shared and services that cannot be shared are mixed. The reason is that the configuration and settings required for the OS, MW, and the like may not be uniform when it is assumed that various services operate. Depending on the service, there are various sharing models with different degrees of module sharing, such as cases where module sharing is not allowed at all, cases where sharing up to hardware by VM (Virtual Machine) is allowed, and cases where sharing up to MW is allowed .

  In this field, Patent Document 1 receives software configuration information of a home network server, creates a list of download program modules, and automatically updates embedded software in a state in which the dependency between program modules is taken into consideration. The module providing server to be executed is described.

  Patent Document 2 describes a system that groups components based on the commonality of dependencies in order to provide a common operating system function set for supporting various operating system configurations.

  Japanese Patent Application Laid-Open No. 2004-133620 describes setting a package having a dependency relationship with a package specified by an installation execution instruction as an installation target.

JP 2005-135187 A JP 2005-235221 A JP 2010-176295 A

  However, with the technique described in the above-mentioned patent document, it has not been possible to deploy modules with a degree of sharing according to the requirements of service owners and the like.

  Therefore, an object of the present invention is to provide a module deployment device, a deployment method, and a deployment program that can deploy modules with an appropriate degree of sharing according to requirements and the like in order to solve the above-described problems.

  In order to achieve such an object, the present invention provides a first module as a designated module and a module in a lower layer of a module in an n-th layer (n is a natural number) as a module in an n + 1-th layer. A module configuration information storage means for storing information on a module string composed of modules up to the lowest layer that is a module without the lower layer module, and information on related modules that are modules of the peer layer of each module; and a server PF configuration information storage means for storing information on modules deployed in a device and sharability information indicating whether other modules can be deployed in an upper layer of the modules deployed in the server device for each module , Module row information obtained from the module configuration information storage means, and the P Comparing the information of the modules deployed on the server device acquired from the configuration information storage means, among the plurality of server devices, at least some of the modules that continue in the upper direction from the lowest layer of the module row are A server module in which a common module (matching module) is deployed and the sharability information of the highest layer of the matching module is deployable is searched, and the module is more than the module deployed in the retrieved server apparatus. Provided is a module deployment device comprising provisioning means for outputting information on an upper layer module in the column and information on the related module of the upper layer module.

  In the present invention, the designated module is a first layer, and a module in a lower layer of a module in the nth layer (n is a natural number) is a module in the (n + 1) th layer. Information of the module sequence obtained from the module configuration information storage means for storing information on the module sequence consisting of the modules up to the lowest layer which is a module having no module and information on the related module which is a module of the peer layer of each module And PF configuration information that stores information on modules deployed in the server device, and shareability information indicating whether other modules can be deployed in higher layers of the modules deployed in the server device for each module A plurality of the servers are compared with the module information acquired from the storage means. A module (matching module) in which at least a part of modules that are continuous in the upper direction from the lowest layer in the module row is shared, and the shareability information of the highest layer of the module can be deployed. A provisioning step of searching for a server device that is configured and outputting information on the upper layer module and information on the related module of the higher layer module in the module row than the module deployed in the searched server device A module deployment program for causing a computer to execute the above is provided.

  In the present invention, the designated module is a first layer, and a module in a lower layer of a module in the nth layer (n is a natural number) is a module in the (n + 1) th layer. Information of the module sequence obtained from the module configuration information storage means for storing information on the module sequence consisting of the modules up to the lowest layer which is a module having no module and information on the related module which is a module of the peer layer of each module And PF configuration information that stores information on modules deployed in the server device, and shareability information indicating whether other modules can be deployed in higher layers of the modules deployed in the server device for each module A plurality of the servers are compared with the module information acquired from the storage means. A module (matching module) in which at least a part of modules that are continuous in the upper direction from the lowest layer in the module row is shared, and the shareability information of the highest layer of the module can be deployed. Module deployment that searches for a server device that is configured, and outputs information on the upper layer module and information on the related module of the higher layer module in the module row than the module deployed on the retrieved server device Provide a method.

  The present invention can provide a module deployment device, a deployment method, and a deployment program that can deploy a module with an appropriate degree of sharing according to requirements and the like.

It is a block diagram which shows an example of a structure of 1st Embodiment. It is a figure which shows an example of the hardware constitutions of each apparatus. It is a flowchart for demonstrating an example of operation | movement of 1st Embodiment. 5 is a diagram illustrating an example of information included in module configuration information 200. FIG. 3 is a diagram illustrating an example of a dependency tree 300. FIG. 4 is a diagram illustrating an example of a configuration of PF configuration information 400. FIG. It is a flowchart which shows an example of the procedure which selects the deployment object server in 1st Embodiment. It is a flowchart which shows an example of the procedure which selects the deployment object server of 2nd Embodiment. It is a flowchart which shows an example of the procedure which selects the deployment object server of 3rd Embodiment. It is a figure which shows an example which expressed the module structure information 200 with the tree. It is a figure which shows an example of a platform structure. It is a figure which shows an example of a platform structure. It is a block diagram which shows an example of a structure of 4th Embodiment.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In all the drawings, the same components are denoted by the same reference numerals, and description thereof will be omitted as appropriate.

  Each unit constituting the device of each embodiment includes a control unit, a memory, a program loaded in the memory, a storage unit such as a hard disk for storing the program, a network connection interface, and the like. Realized by a combination of And unless there is particular notice, the realization method and apparatus are not limited.

  In addition, the block diagram used in the description of each embodiment shows not a hardware unit configuration but a functional unit configuration. These functional blocks are realized by any combination of hardware and software. In addition, in these drawings, the components of each embodiment may be described as being realized by one physically coupled device, but unless otherwise specified, the means for realizing the same is not limited thereto. Not. That is, the components, devices, systems, and the like described in each embodiment may be realized by a plurality of devices by connecting two or more physically separated devices in a wired or wireless manner. Further, although there are cases where each component is described as two or more devices that are physically separated, the realization means is not limited to this unless otherwise specified. That is, each component, device, system, and the like of each embodiment may be realized by arbitrarily combining hardware and software so as to be realized by one physically coupled device.

  FIG. 2 is a diagram illustrating an example of a hardware configuration of each device configuring the deployment system A and the like of FIG. For example, each device includes a control unit 1000, a communication IF (interface) 2000 for network connection, a memory 3000, and a storage device 4000 such as a hard disk for storing programs and the like.

  The control unit 1000 includes a CPU (Central Processing Unit) and the like, and controls an entire apparatus by operating an OS (Operating System). For example, the control unit 1000 can store a program or the like in a memory 3000 from a recording medium mounted on a drive device or the like. Data is read out, and various processes are executed in accordance with the data.

  The recording device 4000 is, for example, an optical disk, a flexible disk, a magnetic optical disk, an external hard disk, a semiconductor memory, etc., and records a computer program so that the computer can read it. The computer program may be downloaded from an external computer (not shown) connected to the communication network. Here, unless otherwise specified, the communication network may be the Internet, a LAN (Local Area Network), a public line network, a wireless communication network, or a network constituted by a combination thereof.

<Embodiment 1>
Next, a first embodiment of the invention will be described in detail with reference to the drawings.

[Overview]
Referring to FIG. 1, the deployment system A according to the first embodiment of the present invention includes a development environment input unit 1, an external repository 2, a deployment device 100, and a data center 10.

  First, the outline of the operation of the present embodiment will be described with reference to FIG.

  The development environment input unit 1 inputs the developed application module (AP module) and module configuration information 200 input from the outside to the deployment device 100.

  The external repository 2 is a repository that stores a platform module (PF module) and related modules. The AP module, PF module, and related modules are collectively referred to as modules.

  The deployment apparatus 100 appropriately collects PF modules on which the AP modules depend from the external repository 2 based on the information input from the development environment input unit 1 and the module deployment status to the server of the data center 10 to obtain data. The AP module input from the development environment input unit 1 is deployed on the center 10.

[Constitution]
Next, the internal configuration of the deployment device 100 will be described in detail with reference to FIG. The deployment apparatus 100 includes a PF (Platform) configuration management unit 101, a PF configuration information storage unit 102, a module management unit 103, a module repository 104, a module configuration management unit 105, a module configuration information storage unit 106, an automatic multi-layer provisioning unit 107, Capacity planning unit 108, AP (Application) module deployment unit 109, related module deployment unit 110, MW (Middleware) deployment unit 111, OS (Operating System) deployment unit 112, VM (Virtual Machine) deployment unit 113, sharing allowable granularity The receiver 114 is configured.

  The PF configuration management unit 101 includes, among servers arranged in the data center 10, PF configuration information 400 indicating which module is deployed on which server, a management device (not shown) that manages each server, each server, and the like. And is recorded in the PF configuration information storage unit 102.

  With reference to FIG. 6, an example of the PF configuration information 400 of the server in the data center 10 at a certain time point managed by the PF configuration information storage unit 102 will be described. The PF configuration information 400 includes server information 410, VM information (virtual machine information) 420, MW information (middleware information) 430, and AP information (application information) 440. The server information 410 holds a list of server information (server information list) 411 in the data center 10. Each record in the list includes a server ID (Identification) for uniquely identifying a server in the data center 10, property information indicating a server specification, and information indicating whether sharing is possible (sharability information). A specific example of shareability information will be described later.

  In the example of FIG. 6, the records 412, 413, and 414 of the server information list 411 refer to the VM information lists 421, 424, and 425 for each server in the VM information 420, respectively. That is, the record 412 of the server ID “SV_a” in the server information list 411 refers to the VM information list 421, the record 413 refers to the VM information list 424, and the record 414 refers to the VM information list 425. These VM information lists show a list of information on virtual machines (VM images) deployed on the servers indicated by the respective server IDs.

  Each record in the VM information list includes a VM_ID that uniquely identifies a deployed virtual machine in the data center 10, OS information indicating the type of OS installed in the virtual machine, and information indicating whether sharing is possible. Each record of the VM information 420 refers to a list of MW information for each VM of the MW information 430. In the example of FIG. 6, the record 422 of the VM_ID “VM_a” in the VM information list 421 refers to the MW information list 431, and the record 423 refers to the MW information list 434. These MW information lists show a list of information on the MW modules deployed on the virtual machines indicated by the respective VM_IDs.

  Each record in the list of MW information includes MW_ID for uniquely identifying the deployed MW module in the data center 10, the type of MW, and information indicating whether sharing is possible. Each record in the MW information list refers to the AP information list for each MW in the AP information 440. In this example, the record 432 of the MW_ID “MW_a” in the MW information list 431 refers to the AP information list 441, and the record 433 refers to the AP information list 442. These AP information lists show a list of information of AP modules deployed on the MW module indicated by each MW_ID.

  Each record in the AP information list includes an AP_ID for uniquely identifying a deployed AP module in the data center 10 and its property information.

  When “No” is specified in the information indicating whether or not a certain module can be shared by an operator or the like, this indicates that a module in a layer higher than that module is not newly deployed in that module. In this example, the server identified by the server ID “SV_b” from the records 413 and 424 is occupied by the virtual machine identified by the VM_ID “VM_d”, and a higher layer module is newly deployed in the virtual machine. Never happen. Further, the virtual machine identified by the VM_ID “VM_b” from the records 423 and 434 is occupied by the MW module identified by the MW_ID “MW_c”, and a module in a higher layer than the MW module is newly deployed in the MW module. It will never be done. Specific examples will be described later with reference to FIGS.

  The module management unit 103 records the PF modules such as the AP module acquired from the development environment input unit 1, the middleware (MW) module acquired from the external repository 2, and the OS module in the module repository 104.

  The module configuration management unit 105 records the module configuration information 200 of the AP module acquired from the development environment input unit 1 in the module configuration information storage unit 106.

  The structure of the module configuration information 200 managed by the module configuration information storage unit 106 will be described with reference to FIG. The module configuration information 200 includes a module name (own module name) 201 indicating the name of the module, and a dependency indicating a module name of a module that is a lower layer immediately below this module, which is a prerequisite for the module to operate on a server or the like. It consists of a module name 205 and a related module name 206 indicating the module name of the module of the peer layer corresponding to the related library necessary for the operation of this module. Here, a layer refers to each layer of the program hierarchical structure, a layer that uses a certain layer is called an upper layer of the layer, and conversely, a layer that is used is called a lower layer. For example, the MW layer is an upper layer of the OS layer, and the OS layer is a lower layer of the MW layer. A layer in the same hierarchy as the module, such as a related library necessary for the operation of the module, is referred to as a peer layer of the module layer. In addition to these pieces of information, the module configuration information 200 further includes a module package 202 that indicates a reference to the module body in a state before the module is deployed, a deployment descriptor 203 that indicates settings at the time of module deployment, and the module package. It may be configured from a packaging 204 indicating a method of performing deployment according to the above.

  The module configuration information 200 is also defined for other modules such as other PF modules and other AP modules referenced from the dependency module name 205 and the related module name 206. Here, the PF module specified by the dependency module name 205 of the module configuration information 200 of the AP module input from the development environment input unit 1 is referred to as a specified module. That is, the designated module is the PF module in the highest layer among the PF modules on which the AP module input from the development environment input unit 1 depends.

  The automatic multilayer provisioning unit 107 can acquire the dependency of the entire module to be deployed on the server necessary for the operation of the AP module by connecting the module configuration information 200. Information indicating this dependency is referred to as a dependency tree 300. An example of the dependency tree 300 is shown in FIG.

  As shown in FIG. 5, the dependency tree 300 has an AP module as a start point (uppermost layer, uppermost layer) and a module for which the dependency module name 205 of the module configuration information 200 is not defined as an end point (lowermost layer, highest layer). (Lower layer) is tree-like information expressing the hierarchical structure of the related module of the peer layer and the PF module of the lower layer based on the dependency module name 205 and the related module name 206.

  The dependency tree 300 is generated by the automatic multilayer provisioning unit 107 with reference to the dependency module name 205 and the related module name 206 of the module configuration information 200, for example. For example, the automatic multi-layer provisioning unit 107 acquires the own module name (referred to as module A) of the module configuration information 200 (referred to as module configuration information A) of the input AP module. Next, the automatic multilayer provisioning unit 107 acquires the module configuration information 200 (referred to as module configuration information B) of the module set in the dependency module name 205 of the module configuration information A from the module configuration information storage unit 106. To do. The automatic multi-layer provisioning unit 107 acquires the own module name (referred to as module B) set in the module configuration information B, and sets the module B in the lower layer of the module A. That is, the automatic multi-layer provisioning unit 107 uses n as a natural number and a module configuration in which the module set in the dependency module name 205 of the module configuration information 200 of the n-th layer module that is the n-th layer is its own module name. It is possible to acquire the platform dependency up to the end point of the dependency tree 300 by repeating the procedure of setting the module in which the information 200 is defined as the (n + 1) th layer module and sequentially setting the lower layer modules. it can.

  Similarly, the automatic multilayer provisioning unit 107 can acquire the dependency of the related module of the peer layer in the same manner as described above with reference to the related module name 206. The dependency tree 300 may be generated by means other than the automatic multilayer provisioning unit 107. The generated dependency tree 300 is stored in a storage unit such as the module configuration information storage unit 106. A linked list described later may also be stored in a storage unit such as the module configuration information storage unit 106.

  The automatic multilayer provisioning unit 107 inquires the capacity planning unit 108 about how much resources are allocated to the server. The automatic multilayer provisioning unit 107 is necessary for AP module deployment using the related module deployment unit 110, the MW deployment unit 111, the OS deployment unit 112, and the VM deployment unit 113 according to the resource allocation acquired from the capacity planning unit 108. A PF module is deployed in the data center 10. Thereafter, the automatic multilayer provisioning unit 107 deploys the corresponding AP module on the server via the AP module deployment unit 109. Here, deployment refers to (1) program transmission, (2) configuration information file change, etc. (3) generation of files required for the operation of deployed modules (4) display of work request messages to the administrator , Response input, etc. Generally, regarding the related modules, the occurrence frequency of (2) to (3) is smaller than that of the PF module.

  The capacity planning unit 108 performs capacity planning of the servers in the data center 10 based on the information of the dependency tree 300 and the service requirements input by the users, and multiplexly deploys AP modules and the like on how many servers. Outputs information (deployment information). Specific methods of capacity planning include a method of estimating how much resources are required from a standard system model measured in advance and a method of estimating the degree of resources by simulation, but the method is not particularly limited. The capacity planning unit 108 can perform capacity planning using a known method. The service requirement is information indicating a service condition requested by a user or the like (time zone during which the service is provided, response time, other service level, etc.), and is stored in the deployment apparatus 100 (not shown) for each service. The capacity planning unit 108 may acquire a service request from the storage unit at a predetermined timing and use it at the time of capacity planning.

  The AP module deploying unit 109, the related module deploying unit 110, the MW deploying unit 111, the OS deploying unit 112, and the VM deploying unit 113 are respectively connected to the server to which each module is deployed in accordance with an instruction from the automatic multi-layer provisioning unit 107. Modules, related modules, MW modules, OS modules, and VM images are deployed.

  The sharing allowable granularity receiving unit 114 receives the sharing allowable granularity designated by the operator according to the intention of the service owner or the like. The sharing allowable granularity is a module lower than the module layer set as the sharing allowable granularity among the modules deployed by the operator or the like on the server in the data center 10 using the deployment device 100. This is information indicating that it can be shared by other modules in the upper layer and the AP module. That is, the module layer immediately below the module layer for which the sharing allowable granularity is set is an upper limit layer that can be shared between the module that the automatic multi-layer provisioning unit 107 intends to deploy on the server and another AP module. The sharing allowable granularity is set by an operator or the like according to the requirements of the service owner, for example. The data center 10 may be prepared (provisioned) in advance with a resource in which a lower layer PF module below the MW module is deployed on the assumption that the AP module is deployed. The sharing allowable granularity receiving unit 114 uses, for example, a service owner as to whether only a server provisioned in this way is a module deployment destination or whether a server in which another AP module is deployed may be a module deployment destination. The sharing allowable granularity can be set to specify according to the case.

[Operation]
Next, the operation of AP deployment (P1000) by the automatic multilayer provisioning unit 107 in the present embodiment will be described in detail with reference to the flowchart of FIG.

  The automatic multi-layer provisioning unit 107 acquires the AP module to be deployed and the module configuration information 200 via the module management unit 103 and the module configuration management unit 105 (S1001). The automatic multilayer provisioning unit 107 acquires the dependency tree 300 based on the dependency module name 205 and the related module name 206 included in the module configuration information 200 (S1002). The automatic multilayer provisioning unit 107 transmits the dependency tree 300 to the capacity planning unit 108. The capacity planning unit 108 performs capacity planning based on the information of the dependency tree 300 and the service requirements input from the outside, and determines how many AP modules are to be deployed in multiple (S1003). As a result of this step, deployment information indicating how many servers the AP module is deployed in multiple is output.

  Next, the automatic multi-layer provisioning unit 107 receives the deployment information from the capacity planning unit 108, and deploys based on the deployment information, the dependency tree 300, and the sharing allowable granularity acquired from the sharing allowable granularity receiving unit 114. The previous server is selected (S1004). A detailed procedure for this step will be described later with reference to FIG. As a result of this step, the automatic multi-layer provisioning unit 107 obtains information about servers to be deployed and which PF modules need to be deployed for each of those servers. If there are not enough servers that can be deployed (Y in S1005), the automatic multilayer provisioning unit 107 outputs an error message indicating that the deployment is impossible (S1006), and ends the series of processes.

  On the other hand, if there is a server that can be deployed (N in S1005), the automatic multi-layer provisioning unit 107 depends on the server via the VM deployment unit 113 if there is a server to which the VM image should be deployed. The VM image specified by the sex tree 300 is deployed (S1007). Next, the automatic multi-layer provisioning unit 107 refers to the dependency module name 205, and if there is a server on which an OS module that is a higher layer of the VM image is to be deployed, the automatic multilayer provisioning unit 107 passes the OS deployment unit 112 to the server. Then, the OS module specified by the dependency tree 300 is deployed (S1008).

  Subsequently, the automatic multi-layer provisioning unit 107 refers to the dependency module name 205, and when there is a server to which the MW module that is the upper layer of the OS module exists, the automatic multi-layer provisioning unit 107 sends the MW module to the server via the MW deployment unit 111. The MW module specified in the dependency tree 300 is deployed (S1009).

  Next, the automatic multi-layer provisioning unit 107 refers to the related module name 206, and when there is a server to which the related module is to be deployed, the related module at the end of the related module name 206 shown in the dependency tree 300. In order, the related modules are deployed to the server via the related module deployment unit 110 (S1010). Next, when there is a server to which the AP module that is the starting point of the dependency tree 300 is to be deployed, the automatic multilayer provisioning unit 107 deploys the AP module to the server via the AP module deployment unit 109. (S1011).

  Finally, the automatic multilayer provisioning unit 107 notifies the PF configuration management unit 101 of the deployment status up to this point, and reflects it in the PF configuration information 400 stored in the PF configuration information storage unit 102 (S1012). In this step, the automatic multi-layer provisioning unit 107 sets the value of the sharing permission / prohibition information in the PF configuration information 400 of the module in the lower layer than the module for which the sharing permission granularity is set to “Yes”. The value of the shareability information in the PF configuration information 400 of the module higher than the set module is set to “No”. This completes a series of AP deployments.

  Next, an example of detailed operation of step S1004 will be described with reference to the flowchart of FIG. In this procedure, first, the automatic multilayer provisioning unit 107 acquires a list (linked list) obtained by extracting only the module configuration information 200 of the PF module from the dependency tree 300 (S2001). The automatic multilayer provisioning unit 107 acquires the sharing allowable granularity from the sharing allowable granularity receiving unit 114 (S2002).

  Next, the automatic multi-layer provisioning unit 107 selects the PF of the highest layer of the PF module indicated by the linked list deployed in the deployment destination server among the servers where all the PF modules indicated by the linked list are deployed. The server whose module shareability information is “Yes” is searched from the PF configuration information 400 stored in the PF configuration information storage unit 102 (S2003).

  Here, the automatic multi-layer provisioning unit 107 stores the PF configuration information when the module of the highest layer in the linked list at that time is a module of a higher layer than the module indicated by the sharing allowable granularity (Y in S2004). With reference to the unit 102, a server in which another AP module is already deployed may be excluded from the search result (S2005).

  Next, the automatic multi-layer provisioning unit 107 compares the PF modules included in the current linked list with the PF modules included in the dependency tree 300 to obtain a list of PF modules to be deployed on the server in the previous step. Are added to the result list together with the server name found as a result (S2006). Specifically, the automatic multilayer provisioning unit 107 compares the PF module included in the current linked list with the PF module included in the dependency tree 300, and is included in the dependency tree 300. PF modules not included in the list are extracted. For example, the automatic multi-layer provisioning unit 107 sequentially compares whether or not each PF module is included in the current linked list from the module of the dependency tree 300 layer toward the lower layer. The missing PF module is extracted. The automatic multilayer provisioning unit 107 sets the extracted PF module as a PF module to be deployed on the server, and associates the PF module to be deployed on the server with the server name discovered as a result of the previous step. Add to result list. The result list is a list that stores a server name and a PF module deployed to the server in association with each other, and is stored in a storage unit (not shown) by the automatic multilayer provisioning unit 107. At this time, if the number of servers included in the result list has reached the requested number of servers (Y in S2007), the automatic multilayer provisioning unit 107 lists the modules to be deployed on the searched server together with the result list. May be output (S2008).

  On the other hand, if the number of servers in the result list has not yet reached the requested number of servers (N in S2007), the automatic multilayer provisioning unit 107 deletes the PF module in the highest layer in the linked list (S2009). If there is no PF module in the linked list and there is no PF module that can be deleted (N in S2010), the automatic multilayer provisioning unit 107 outputs that there are not enough deployment destination servers (S2011), and ends the procedure. To do.

  If the PF module still remains in the linked list (Y in S2010), the procedure from step S2003 is repeated.

  Next, the outline of the operation of the present embodiment will be described with reference to FIGS.

  FIG. 10 shows how the configuration information 500 of an application AP1 501 is expressed as a tree structure. In the application AP 1 501, MW 2 502 is set as the dependent module name 205, and Lib A 503 and Lib B 504 are set as the related module name 206. Further, in the configuration information 510 of MW2, MW1 512 is set as the dependency module name 205. Furthermore, in the configuration information 520 of the MW1 520, OS1 522 is set as the dependency module name 205. Further, it is assumed that MW2 502 is set as a sharing allowable granularity by an operator or the like.

  It is assumed that the automatic multilayer provisioning unit 107 acquires the deployment information indicating that this configuration is installed in four virtual environments from the capacity planning unit 108. FIG. 11 shows the platform configuration in the data center 10. As shown in FIG. 11, MW1 deployed in the server 701 and higher layer modules have “NO” sharability information, and the VM deployed in the server 703 and higher layer layers. The module has “NO” shareability information.

  When the automatic multilayer provisioning unit 107 first extracts the linked list from the configuration information of the AP1, it becomes MW2, MW1, and OS1. The automatic multilayer provisioning unit 107 searches a server in the data center 10 for a server having all the configurations of the linked list MW2, MW1, and OS1.

  Among the servers in the data center 10, the servers having a virtual environment that satisfies this configuration are two servers, the server 701 and the server 702, and the automatic multilayer provisioning 107 searches for these two servers. However, since the MW2 shareability information of the server 701 is “NO”, at this stage, only the server 702 is a candidate for the deployment destination server, and the AP1 module is the deployment target.

  Since this alone is less than the required four units, the automatic multilayer provisioning 107 deletes the module of the highest layer in the linked list to make MW1 and OS1. Servers having a virtual environment that satisfies this configuration are a server 701 and a server 704. However, since the shareability information of MW1 701 is “NO”, only the server 704 is added at this stage.

  Since there are not yet two deployment destination server candidates, the automatic multi-layer provisioning 107 further deletes the module of the highest layer in the linked list to be OS1. As the virtual environment satisfying this configuration, 701 and 706 correspond, and the OS1 layer has the shareability information “Yes”. Therefore, the automatic multi-layer provisioning 107 selects the initial request by selecting 701 and 706. Can be selected.

  Next, the automatic multilayer provisioning 107 sequentially deploys the PF modules to the selected server. First, the automatic multilayer provisioning 107 deploys MW1 to the server 701 and the server 706 via the MW deployment unit 111. Subsequently, the automatic multilayer provisioning 107 deploys MW2 to the server 701, the server 704, and the server 706 via the MW deployment unit 111. Next, the automatic multi-layer provisioning 107 deploys the related modules LibA and LibB to the server 701, the server 702, the server 704, and the server 706 via the related module deployment unit 110. Further, the automatic multilayer provisioning 107 deploys the AP1 module to these servers via the AP module deployment unit 109.

  Since the allowable sharing granularity is MW2, the automatic multi-layer provisioning 107 sets “No” to the sharing permission / inhibition information of MW2 or more of the newly deployed module. The state of the server after module deployment is shown in FIG.

  With such a configuration, the deployment apparatus 100 according to the present embodiment is configured such that a module that is a premise of the application operation and a related module such as a library based on the PF module deployment status to the server in the data center 10 and the shareability information. Applications can be deployed with integrated provisioning.

<Embodiment 2>
Next, a second embodiment of the present invention will be described in detail with reference to the drawings. The present embodiment is different from the first embodiment in that the deployment apparatus 100 selects a deployment destination server so that other AP modules do not share the platform as much as possible in order to ensure the performance of each AP module. Further, the shared permissible granularity receiving unit 114 is different from the first embodiment in that it also receives a shared permissible granularity indicating whether the AP module deployed in the server permits sharing of resources such as MW modules with other AP modules. .

  The selection of the placement server in step S1004 of FIG. 3 in the present embodiment will be described using the flowchart of FIG.

  The stage from the start of selection P2100 of the arrangement server in FIG. 8, that is, the procedure from step S1004 in FIG. 3 described above will be described. In this procedure, first, the automatic multi-layer provisioning unit 107 determines how many other AP modules are deployed in order to distribute the AP module deployment as much as possible when extracting the deployment target server candidates. The value of the shared number counter to be determined is initialized to 1 (S2101).

  The automatic multilayer provisioning unit 107 acquires a linked list from the dependency tree 300 (S2102). The automatic multilayer provisioning unit 107 acquires the sharing allowable granularity of the AP module to be deployed from the sharing allowable granularity receiving unit 114 (S2103).

  Next, the automatic multi-layer provisioning unit 107 has deployed all the PF modules indicated by this linked list from the PF configuration information storage unit 102 via the PF configuration management unit 101, and is the highest layer that is the direct deployment destination. A server whose shareability information of the PF module (the PF module of the highest layer of the PF module shown in the current linked list deployed in the deployment destination server) is “Yes” is searched (S2104). The automatic multi-layer provisioning unit 107 extracts a server in which the number of AP modules already deployed is less than the value of the shared number counter among the servers extracted in step S2104 (S2105). Here, when the module of the highest layer of the linked list at that time is a module of a higher layer than the module for which the sharing allowable granularity is set (Y in S2106), the automatic multi-layer provisioning unit 107 already has another The server on which the AP module is deployed may be excluded from the search result (S2107).

  Next, the automatic multi-layer provisioning unit 107 compares the PF modules included in the current linked list with the PF modules included in the dependency tree 300 to obtain a list of PF modules to be deployed on the server in the previous step. The server name found as a result is added to the result list (S2108). If the number of servers included in the result list has reached the requested number of servers at this time (Y in S2109), the automatic multilayer provisioning unit 107 outputs the result list at that time (S2110), and performs the procedure. finish.

  If the number of servers in the result list has not yet reached the requested number of servers (N in S2109), the automatic multilayer provisioning unit 107 deletes the PF module in the highest layer in the linked list (S2111). If the PF module still remains in the linked list (Y in S2112), the automatic multilayer provisioning unit 107 repeats the procedure from step S2104. If there is no PF module that can be deleted in the linked list (N in S2112), the automatic multilayer provisioning unit 107 increments the value of the shared number counter by 1 (S2113).

  If the value of the shared number counter does not exceed the preset upper limit value (N in S2114), the automatic multilayer provisioning unit 107 repeats the procedure from step S2102. When the value of the shared number counter exceeds the upper limit value (Y in S2114), the automatic multilayer provisioning unit 107 outputs that there are not enough deployment destination servers (S2115), and ends the procedure.

  With such a configuration, the deployment apparatus 100 according to the present embodiment performs a module from a premise of application operation to a related module such as a library according to the PF module deployment status to the server in the data center 10 and the shareability information. Can be provisioned and deployed to deploy applications. Further, the deployment apparatus 100 can select a deployment destination server so that other AP modules do not share the platform as much as possible in order to ensure the performance of each AP module.

<Embodiment 3>
Next, a third embodiment of the present invention will be described in detail with reference to the drawings. The present embodiment is different from the first embodiment in that the deployment destination server is selected so as to share the platform with other AP modules as much as possible in order to deploy modules on fewer servers. The configuration is the same as that of the second embodiment.

  The selection of the placement server in step S1004 of FIG. 3 in the present embodiment will be described using the flowchart of FIG.

  The stage at which placement server selection P2200 starts, that is, the procedure after step S1004 in FIG. 3 described above will be described. In this procedure, first, when the automatic multi-layer provisioning unit 107 extracts candidates for the deployment target server, the number of shares that serves as a guideline for selecting a server in which more than one AP module is deployed in order to increase the degree of sharing. The counter value is initialized to a predetermined maximum value (S2201).

  The automatic multilayer provisioning unit 107 acquires a linked list from the dependency tree 300 (S2202). Next, the automatic multi-layer provisioning unit 107 acquires the sharing allowable granularity of the AP module to be deployed from the sharing allowable granularity receiving unit 114 (S2203).

  Next, the automatic multi-layer provisioning unit 107 has deployed all the PF modules shown in the linked list from the PF configuration information storage unit 102 via the PF configuration management unit 101, and is the highest layer that is the direct deployment destination. A server whose shareability information of the PF module is “Yes” is searched (S2204). The automatic multilayer provisioning unit 107 extracts, from the servers extracted in step S2204, a server in which the number of already deployed AP modules is equal to or greater than the value of the shared number counter (S2205). Here, when the module of the highest layer of the linked list at that time is a module of a higher layer than the module for which the sharing allowable granularity is set (Y in S2206), the automatic multi-layer provisioning unit 107 The server on which the module is deployed may be excluded from the search result (S2207).

  Next, the automatic multi-layer provisioning unit 107 compares the PF modules included in the current linked list with the PF modules included in the dependency tree 300 to obtain a list of PF modules to be deployed on the server, as a result of the previous step. It adds to a result list with the discovered server name (S2208). If the number of servers included in the result list has reached the requested number of servers at this time (Y in S2209), the automatic multilayer provisioning unit 107 returns the result list at that time (S2210), and performs the procedure. finish.

  If the number of servers in the result list has not reached the requested number of servers (N in S2209), the automatic multilayer provisioning unit 107 deletes the PF module in the highest layer in the linked list (S2211). When modules in the linked list still remain (Y in S2212), the automatic multilayer provisioning unit 107 repeats the procedure from step S2204.

  When there is no PF module that can be deleted in the linked list (N in S2212), the automatic multilayer provisioning unit 107 decreases the value of the shared number counter by 1 (S2213). If the value of the shared number counter is 0 or more (N in S2214), the automatic multilayer provisioning unit 107 repeats the procedure from step S2202.

  When the value of the shared number counter becomes a negative number (Y in S2214), the automatic multilayer provisioning unit 107 returns that there are not enough deployment destination servers (S2215) and ends the procedure.

  With such a configuration, the deployment apparatus 100 according to the present embodiment performs a module from a premise of application operation to a related module such as a library according to the PF module deployment status to the server in the data center 10 and the shareability information. Can be provisioned and deployed to deploy applications. Further, the deployment apparatus 100 can select the deployment destination server so as to share the platform with other AP modules as much as possible in order to deploy the modules to fewer servers.

<Embodiment 4>
Next, a fourth embodiment of the present invention will be described in detail with reference to the drawings.

  The deployment apparatus 100 according to the present embodiment includes an automatic multilayer provisioning unit 107, a module configuration information storage unit 106, and a PF configuration information storage unit 102.

  The module configuration information storage unit 106 is a module composed of modules from the first layer to the lowest layer, with the designated module being the first layer and the module of the (n + 1) th layer being the module of the lower layer of the module of the nth layer. The column information and the information of the related module that is the module of the peer layer of each module are stored as the module configuration information 200, for example.

  The PF configuration information storage unit 102 stores sharability information indicating sharability of modules deployed in the server.

  The automatic multilayer provisioning unit 107 selects a server from the plurality of servers based on the information on the module row, the sharability information, and the information on the modules deployed from the outside acquired from the outside. In the module row, the module information of the upper layer and the information of the related module of the module of the upper layer are output in the module row than the modules deployed in the server.

  With such a configuration, the deployment apparatus 100 according to the present embodiment can change from a module that is a prerequisite for application operation to a related module such as a library according to the PF module deployment status to the server in the data center 10 and information on whether or not to share. Applications can be deployed with integrated provisioning.

  While the present invention has been described with reference to the embodiments, the present invention is not limited to the above embodiments. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

DESCRIPTION OF SYMBOLS 1 Development environment input part 2 External repository 10 Data center 100 Deployment apparatus 101 PF configuration management part 102 PF configuration information storage part 103 Module management part 104 Module repository 105 Module configuration management part 106 Module configuration information storage part 107 Automatic multilayer provisioning part 108 Capacity planning unit 109 AP module deployment unit 110 Related module deployment unit 111 MW deployment unit 112 OS deployment unit 113 VM deployment unit 114 Shared permissible granularity reception unit

Claims (10)

The designated module is the first layer, and the module in the lower layer of the module in the nth layer (n is a natural number) is the module in the n + 1th layer. Module configuration information storage means for storing information on a module string composed of modules up to the lower layer and information on related modules that are modules of the peer layer of each module;
PF configuration information storage means for storing information on modules deployed in the server device and sharability information indicating whether or not other modules can be deployed in higher layers of the modules deployed in the server device for each module When,
The module sequence information acquired from the module configuration information storage means and the module information provided in the server device acquired from the PF configuration information storage means are compared, and the module is selected from the plurality of server devices. A server device in which a module (matching module) common to at least a part continuous in the upper direction from the lowest layer of the column is provided, and the shareability information of the highest layer of the matching module is searchable is searched. Provisioning means for outputting higher layer module information and the related module information of the higher layer module in the module row than the module deployed in the searched server device;
A module deployment device comprising: Further, the PF configuration information storage means for storing information on whether or not an application is deployed in the server device;
A share allowable granularity receiving means for receiving a share allowable granularity specifying one layer in the module sequence;
Referring to the PF configuration information storage means, among the plurality of server devices, the module of the highest layer of the matching module is a layer higher than the layer specified by the sharing allowable granularity, and an application The module deployment device according to claim 1, further comprising: the provisioning unit that searches for a server device other than the server device on which the server is deployed. The provisioning means includes
The sharing of the modules of the first layer among the modules arranged in the server device among the plurality of the server devices and all the modules in the module row from the first layer to the lowest layer. When a server device for which availability information can be deployed is searched and the server device does not exist, a degenerate module sequence is generated by sequentially removing modules one by one from the upper layer of the module sequence, and the degenerate module sequence A server device including all modules, wherein a search is made for a server device in which the shareability information of a module in the highest layer of the degenerate module deployed in the server device can be deployed, and the server device searched for the server device The module deployment device according to claim 1 or 2, which performs output.   The provisioning means searches for a server device in which the number of the applications deployed in the server device is less than a predetermined number among the plurality of server devices, and when the server device does not exist, the provisioning unit is predetermined. The module deployment device according to claim 3, wherein the predetermined number is sequentially incremented by 1 until reaching the threshold value, and the server device having the number of modules less than the predetermined number is searched.   The provisioning means searches for a server device having a predetermined number or more of the applications deployed on the server device among the plurality of server devices, and becomes negative if the server device does not exist. The module deployment apparatus according to claim 3, wherein the predetermined number is sequentially decreased by 1 until the number of the modules is searched for the server device. The provisioning means includes
6. The module deployment apparatus according to claim 1, wherein the output upper layer module and the related module are acquired from a module storage unit that stores the modules, and are deployed on the server apparatus. The designated module is the first layer, and the module in the lower layer of the module in the nth layer (n is a natural number) is the module in the n + 1th layer. Information on the module sequence obtained from the module configuration information storage means for storing information on the module sequence consisting of each module up to the lower layer and information on the related module that is a module of the peer layer of each module;
PF configuration information storage means for storing information on modules deployed in the server device and sharability information indicating whether or not other modules can be deployed in higher layers of the modules deployed in the server device for each module Compare with the module information obtained from
Among the plurality of server devices, a module (matching module) that is common to at least a part continuous in the upper direction from the lowest layer of the module row is arranged, and the shareability information of the highest layer of the matching module is The server device that can be deployed is searched, and the information of the upper layer module and the information of the related module of the upper layer module are output in the module row than the module deployed in the retrieved server device. A module deployment program for causing a computer to execute a provisioning step. Receiving a sharing allowed granularity specifying one layer in the module sequence;
Further, referring to the PF configuration information storage means for storing information on whether or not an application is deployed in the server device, the highest layer module of the matching module is higher than the layer specified by the sharing allowable granularity The module deployment program according to claim 7, wherein the computer executes the provisioning step of searching for a server device other than the server device to which the application is deployed, and which is a layer of the application. The designated module is the first layer, and the module in the lower layer of the module in the nth layer (n is a natural number) is the module in the n + 1th layer. Information on the module sequence obtained from the module configuration information storage means for storing information on the module sequence consisting of each module up to the lower layer and information on the related module that is a module of the peer layer of each module;
PF configuration information storage means for storing information on modules deployed in the server device and sharability information indicating whether or not other modules can be deployed in higher layers of the modules deployed in the server device for each module Compare with the module information obtained from
Among the plurality of server devices, a module (matching module) that is common to at least a part continuous in the upper direction from the lowest layer of the module row is arranged, and the shareability information of the highest layer of the matching module is The server device that can be deployed is searched, and the information of the upper layer module and the information of the related module of the upper layer module are output in the module row than the module deployed in the retrieved server device. How to deploy modules. Receiving a sharing allowable granularity specifying one layer in the module sequence;
Further, referring to the PF configuration information storage means for storing information on whether or not an application is deployed in the server device, the highest layer module of the matching module is higher than the layer specified by the sharing allowable granularity The module deployment method according to claim 9, wherein a server device other than the server device on which an application is deployed is searched for.

Images