JP6981124B2 - System construction support device, system construction support method, and program - Google Patents

Description

The present invention relates to a system construction support device for supporting the construction of a system defined by a declarative model, and a system construction support method, and further relates to a program for realizing these.

In recent years, with the development of cloud infrastructure and virtualization infrastructure, it has become possible to use a large amount of computer resources at low cost. Infrastructure as Code (denoted as "IaC") is known as a technique for managing such computer resources.

According to IaC, the system can be defined by the source code, and the system can be built automatically by a tool called an orchestrator. Further, as specific examples of IaC, OpenStack Heat, AWS (Amazon Web Service) Cloud formation, Ansible, etc. are known.

In addition, there is a declarative model as one of the methods of defining a system by IaC. Specifically, the Heat Orchestration Template (HOT) of the OpenStack Heat component, the Cloud Formation Template of the AWS Cloud formation service, etc. In addition, the declarative model exists in the format of playbook in Ansible.

Also, declarative models are generally created by writing source code using a text editor. Further, a method of creating a declarative model using a graphical editor for an orchestrator of a declarative model instead of using a text editor has also been proposed (see, for example, Non-Patent Document 1).

Specifically, Non-Patent Document 1 discloses a declarative model creation tool for AWS Cloud Formation, such as AWS Cloud Formation Designer, in order to change the definition of AWS Cloud Formation. The tool disclosed in Non-Patent Document 1 is a graphical editor, which makes it possible to visually create a declarative model.

However, it is difficult to "correctly" define a declarative model because it is not easy to define the dependencies between the components of the declarative model using any editor. The reason why it is not easy to define the dependency between the components is that the creator of the declarative model does not know (does not know) whether or not there is a dependency between the components, and when creating the declarative model. The point is that the dependency information of the components used for may not be defined correctly.

The "component" is a unit used when creating a declarative model. Examples of the components include general-purpose ones such as a Web server and a DB (Data Base) server, and specific ones such as a Tomcat server and a MySQL server. In addition, the component is expressed as Resource Type in AWS Cloud Formation and Node Type in OASIS TOSCA.

Further, "dependency information" indicates a definition of a requirement required by a certain component and a capability that a certain component can provide. In OASIS TOSCA, it is expressed by the definition of Requirement and Capability of a certain Node Type.

Therefore, in the past, when creating a new declarative model, refer to the sample definition and modify the declarative model, or perform actual provisioning using the created definition and issue Try & Error. It is necessary to repeatedly converge to the "correct" definition. Further, Patent Document 1 discloses a technique of analyzing a created declarative model based on the information of the construction result and further improving the declarative model based on the analysis result.

It is possible to reuse the definition of the declarative model using a conversion tool, but it is impossible to use it as it is, with some exceptions (AWS CFN description can be used in OpenStack). Is. In addition, OASIS TOSCA is trying to add portability to the definition by formulating a standard definition, but even in this case, Try & Error is necessary.

Special Table 2010-532532

Jun Niizaki, 3 outsiders, "Trends in new technology", [online], April 1, 1998, Patent Society, [Search on July 30, 1999], Internet <URL: https: // aws. amazon.com/jp/cloudformation/details/#designer ＞

By the way, according to the technique disclosed in Patent Document 1 described above, the declarative model can be efficiently improved as compared with the case of referring to the sample definition and the case of repeating Try & Error. There is a problem that unintended improvements are made. The reasons are as follows (a) to (f).

(A) The declarative model is improved by focusing on the improvement of the dependency between the components, but the individual components are not improved.
(B) Regarding the instruction generated from the declarative model, the case where an error occurs at the time of actual execution is not considered.
(C) Knowledge for making improvements is not clear.
(D) Patent Document 1 exemplifies the improvement of the load balancing configuration mainly by the performance information, but the intentional knowledge by the creator (editor user) of the declarative model incorporated in the actual system is available. Not built in.
(E) Improvements have been made after registering the declarative model in the repository.
(F) The creator of the declarative model (editor user) should be presented with improvements based on the dependency information at the stage of creating the declarative model, but such presentation has not been made.

An example of the object of the present invention is a system construction support device and system construction support that can solve the above problems and present information for guiding the declarative model in the direction intended by the user in the creation of the declarative model. To provide methods and programs.

In order to achieve the above object, the system construction support device in one aspect of the present invention is a device for supporting the construction of the system defined by the declarative model.
An information gathering unit that collects dependency information that identifies the dependencies of the components that make up the declarative model.
Support information that creates support information to support the modification of the declarative model by applying the dependency information to the prediction model that predicts the influence of the dependency of the component on the declarative model. With the creation department
It is characterized by having.

Further, in order to achieve the above object, the system construction support method in one aspect of the present invention is a method for supporting the construction of the system defined by the declarative model.
(A) A step that collects dependency information that identifies the dependencies of the components that make up the declarative model.
(B) By applying the dependency information to a prediction model that predicts the influence of the dependency of the component on the declarative model, support information for supporting the modification of the declarative model is created. , Steps and
It is characterized by having.

Further, in order to achieve the above object, the program in one aspect of the present invention is a program for supporting the construction of a system defined by a declarative model by a computer.
To the computer
(A) A step that collects dependency information that identifies the dependencies of the components that make up the declarative model.
(B) By applying the dependency information to a prediction model that predicts the influence of the dependency of the component on the declarative model, support information for supporting the modification of the declarative model is created. , Steps and
Is characterized by executing.

As described above, according to the present invention, in creating a declarative model, it is possible to present information for guiding the declarative model in the direction intended by the user.

FIG. 1 is a block diagram showing a schematic configuration of a system construction support device according to an embodiment of the present invention. FIG. 2 is a block diagram showing a specific configuration of the system construction support device according to the embodiment of the present invention. FIG. 3 is a diagram showing an example of local information and public information used in the embodiment of the present invention. FIG. 4 is an explanatory diagram showing an example of the relationship between the definition of the declarative model used in the present embodiment and the components of the system. FIG. 5 is a flow chart showing the operation of the system construction support device according to the embodiment of the present invention. FIG. 6 is a diagram showing the effects obtained by the system construction support device according to the embodiment of the present invention. FIG. 7 is a block diagram showing an example of a computer that realizes the system construction support device according to the embodiment of the present invention.

(Embodiment)
Hereinafter, the system construction support device, the system construction support method, and the program according to the embodiment of the present invention will be described with reference to FIGS. 1 to 7.

[Device configuration]
First, the configuration of the system construction support device according to the embodiment of the present invention will be described with reference to FIG. FIG. 1 is a block diagram showing a schematic configuration of a system construction support device according to an embodiment of the present invention.

The system construction support device 1 in the present embodiment shown in FIG. 1 is a device for supporting the construction of the system 1000 defined by the declarative model. As shown in FIG. 1, the system construction support device 1 includes an information collecting unit 2 and a support information creating unit 3.

The information collecting unit 2 collects dependency information that identifies the dependencies of the components constituting the declarative model. The support information creation unit 3 creates support information for supporting the modification of the declarative model by using the prediction model that predicts the influence of the dependency of the component on the declarative model.

As described above, in the present embodiment, the support information for supporting the modification of the declarative model is created. Therefore, in the creation of the declarative model, the declarative model is guided in the direction intended by the user. Information can be presented.

Subsequently, with reference to FIG. 2, the configuration of the system construction support device 1 in the present embodiment will be described more specifically. FIG. 2 is a block diagram showing a specific configuration of the system construction support device according to the embodiment of the present invention.

As shown in FIG. 2, in the present embodiment, the system construction support device 1 is incorporated in the system 1000 to be supported. Specifically, as will be described later, it is constructed by a program installed in a computer constituting the system 1000.

Further, as shown in FIG. 2, the system 1000 includes a declarative model editing component 100, a model repository 101, a declarative model orchestrator 200, a workflow execution component 300, and an execution environment 400.

The declarative model editing component 100 is a tool that accepts the editing of the declarative model by the user and further outputs the source code such as JSON or YAML in response to the instruction by the user. The declarative model editing component 100 also generally functions as an editor having a GUI (Graphical User Interface).

Further, the declarative model 11 (text information) edited by the declarative model editing component 100 is held in the model repository 101. The model repository 101 is a repository that stores the declarative model 11 in the local environment.

The declarative model orchestrator 200 interprets the declarative model 11 and, depending on the interpretation, transforms the declarative model into an internal interpreter for performing execution processing. Further, this internal interpreter is described as workflow 21 in FIG. 2, and is composed of text information. The workflow 21 is sent to the workflow execution component 300. Further, the declarative model orchestrator 200 is implemented by the TOSCA reference implementation, the OpenStack Heat implementation, and the like.

The workflow execution component 300 specifically indicates the declarative model interpreted by the declarative model orchestrator 200 according to the execution platform for constructing the system 1000, that is, the actual provisioning process is performed according to the workflow 21. To execute. Further, in order to correspond to various execution environments, the workflow execution component 300 is implemented by a plug-in pattern that executes data conversion, API call, etc. corresponding to each execution environment.

The execution environment 400 is an execution environment for operating an actual system. Specific examples of the execution environment 400 include VMware, OpenStack, AWS, and the like. Further, as shown in FIG. 2, in the execution environment 400, the system definition information 401 and the log information 402 are held.

The system definition information 401 is a set of definition information related to the system in the actual operating environment. Examples of the system definition information 401 include a setting file of a Web server application, a kernel parameter of an operating system, and the like. The log information 402 is a set of log information related to the system in the actual operating environment.

Further, the model component repository 700 is a repository for storing components when creating a declarative model. Specifically, the model component repository 700 is called Node Type in OASIS TOSCA, for example.

Further, as shown in FIG. 2, the system 1000 is connected to the Internet 600. The system 1000 accesses the public repository group 601 published on the Internet 600. Examples of public repositories 601 include definition information (GitHub, Docker Hub, etc.) used when using declarative model tools (eg, OpenStack Heat, AWS Cloud Formation, etc.) and IaC tools (eg, Ansible, Docker, etc.). Ansible-Galaxy etc.). These definition information are published on the Internet 600.

Further, as shown in FIG. 2, in addition to the above-described configuration, the system 1000 is open to the public with a declarative model analysis component 510, a workflow execution status analysis component 520, an existing system analysis component 530, and a log analysis component 540. It also has a system configuration information analysis component 550.

These analysis components 510 to 550 are components used for collecting dependency information by the information collecting unit 2. The analysis component 510 to 550 acquires the dependency information (text information) from the information of the user's local environment (local information) and the information published on the Internet (public information), and the acquired dependency information. Is input to the information collecting unit 2.

In addition to the information related to direct dependencies, the information acquired by these analysis components 510 to 550 includes semantic information related to errors, performance monitoring results, and components (for example, a Web server or a DB server). ), Information indicating that it is a best practice, information indicating that it is an anti-pattern, etc. are included.

Further, the local information means the information held in the model repository 101 and the execution environment 400. Further, the public information means the information held in the public repository group 601. FIG. 3 is a diagram showing an example of local information and public information used in the embodiment of the present invention.

Specifically, the declarative model analysis component 510 analyzes the dependency definition of the declarative model 11 stored as local information, and the information (analysis result 51) obtained by the analysis is used as the system construction support device 1. Send to.

The workflow execution status analysis component 520 analyzes an error from the workflow execution result (for example, API return value, log information) using the declarative model by the workflow execution component 300. Errors include errors that occur in the dependency definition. Then, the workflow execution status analysis component 520 sends the information (analysis result 52) obtained by the error analysis to the system construction support device 1.

The existing system analysis component 530 analyzes the definition of the system 1000 based on the general information of the system 1000. Specifically, the existing system analysis component 530 analyzes the information about the system operating on the execution environment 400 and the information about the dependency existing in the base thereof. Then, the existing system analysis component 530 sends the information (analysis result 53) obtained by the analysis of the definition to the system construction support device 1.

The log analysis component 540 analyzes the log (log information 402) held in the system 1000, and sends the information (analysis result 54) obtained by the analysis to the system construction support device 1. Examples of the log in this case include a system log generated on the execution environment 400, a log in an application program, and the like.

Further, the public system configuration information analysis component 550 analyzes the public repository group 601 published on the Internet 600, and sends the information (analysis result 55) obtained by the analysis to the system construction support device 1. Specifically, the public system configuration information analysis component 550 acquires the system definition information 61 and the usage status information 62 from the public repository group 601.

The system definition information 61 is public information regarding a declarative model and a system configuration, and specifically, is text information such as OpenStack Heat Orchestration Template (HOT), AWS Cloud Formation Template, Dockerfile, and playbook. Further, the usage status information 62 is data indicating the usage status of the information published on the Internet 600, for example, the number of downloads, the number of Likes, the number of Folks on GitHub, and the like.

Further, in the present embodiment, the analysis component used is not limited to the above-mentioned analysis component. Depending on the type and number of dependency information collected, the number of analysis components will increase, and the accuracy of dependency complementation will improve.

Further, as shown in FIG. 2, in the present embodiment, the system construction support device 1 further includes a prediction model generation unit 4 and a prediction model storage in addition to the above-mentioned information collection unit 2 and support information creation unit 3. A unit 5 and a dependency bulk data database (DB: Data Base) 6 are provided.

The prediction model generation unit 4 generates a prediction model by performing machine learning using the dependency information for learning and the declarative model for learning as teacher data. Specifically, the prediction model generation unit 4 first extracts features from each of the dependency information for learning and the declarative model for learning. Next, the prediction model generation unit 4 uses the extracted dependency information feature amount and the declarative model feature amount as "generalization", "correctness", "local dependency", and "anti-pattern" for the components. ”Is classified by items such as. Then, the prediction model generation unit 4 associates each feature amount with the classification result, learns these relationships, and generates a prediction model using the learning result. The generated prediction model is stored in the prediction model storage unit 5.

Further, the prediction model generation unit 4 can also perform machine learning by distinguishing the dependency information for learning into local information and public information. This makes it possible to create a more accurate prediction model. This is because when public information is used as the dependency information for learning, the learning effect can be obtained from the widely and generally used overall optimum viewpoint. Further, when local information is used as the dependency information for learning, the learning effect regarding the unique know-how for constructing and operating the system can be obtained.

Furthermore, when local information is used as the dependency information for learning, the prediction model makes it possible to propose system-specific adjustments that are different from the general definition when creating a declarative model. In addition, in this case, it will be possible to define an existing on-premises system without dropping important relationships and operational items in creating a declarative model for migrating to a cloud environment, for example.

In the present embodiment, the information collecting unit 2 acquires the dependency information sent from each of the above-mentioned analysis components 510 to 550 and collects them. The information collecting unit 2 stores the collected dependency information in the dependency bulk data DB 6.

Further, the information collecting unit 2 collects local information from the declarative model analysis component 510, the workflow execution status analysis component 520, the existing system analysis component 530, and the log analysis component 540 among the above-mentioned analysis components. Further, the information collecting unit 2 collects public information from the public system configuration information analysis component 550. Therefore, the information collecting unit 2 can collect the local information and the public information separately, and can attach a label indicating whether the collected information is the local information or the public information to each of the collected information.

In the present embodiment, the support information creation unit 3 creates support information by applying the dependency information stored in the dependency bulk data DB 6 to the prediction model stored in the prediction model storage unit 5. do. Further, when the above-mentioned label is attached to the dependency information collected by the information collecting unit 2, the support information creating unit 3 applies the dependency information to the prediction model together with the attached label. In this case, more appropriate support information that reflects the contents of the label will be created.

Specifically, the support information creation unit 3 infers whether or not the workflow operates according to the created declarative model by using the prediction model. Further, the guessing at this time may be performed by hypothetical reasoning using a prediction model. Then, the support information creation unit 3 creates, as support information, information for supporting the generation, complementation, or correction of the dependency between the constituent elements in the creation of the declarative model. Further, the support information creation unit 3 provides an API in a typical implementation.

Here, with reference to FIG. 4, the relationship between the definition of the declarative model used in the present embodiment and the components of the system will be described. FIG. 4 is an explanatory diagram showing an example of the relationship between the definition of the declarative model used in the present embodiment and the components of the system.

FIG. 4 shows the concept that a declarative model is constructed based on the components. Further, as shown in FIG. 4, the component and the entity in the declarative model are metaphorized as a class definition and its entity object. Further, in the example of FIG. 4, the declarative model and components defined by OASIS TOSCA are shown. OASIS TOSCA is one of the specifications for defining the system configuration (topology) of a business system running on the cloud. Further, in FIG. 4, the dependency relationship of each component is represented by associating "Requirements" and "Capabilities" added to the component with "Hosted On" (see FIG. 3).

[Device operation]
Next, the operation of the system construction support device 1 in the present embodiment will be described with reference to FIG. FIG. 5 is a flow chart showing the operation of the system construction support device according to the embodiment of the present invention. Further, in the present embodiment, the system construction support method is implemented by operating the system construction support device 1. Therefore, the description of the system construction support method in the present embodiment is replaced with the following operation description of the system construction support device 1.

As shown in FIG. 5, first, in the system construction support device 1, the information collecting unit 2 collects the dependency information for specifying the dependency of the components constituting the declarative model (step A1). Further, the information collecting unit 2 stores the collected dependency information in the dependency bulk data DB 6.

Specifically, in the present embodiment, the declarative model analysis component 510, the work row execution status analysis component 520, the existing system analysis component 530, the log analysis component 540, and the public system configuration information analysis component 550 each periodically. Alternatively, the analysis is executed with a specific event as an opportunity, and the analysis result obtained by the analysis is transmitted. The information collecting unit 2 collects the transmitted analysis result as dependency information.

Next, the support information creation unit 3 creates support information by applying the dependency information collected in step A1 to the prediction model stored in the prediction model storage unit 5 (step A2).

Specifically, the support information creation unit 3 declares that the support information is information that allows the user to guide the declarative model in a more correct direction, and that a system configuration that can withstand usability is constructed. Create information that complements the target model.

After that, the support information creation unit 3 presents the support information created in step A2 to the user (step A3). Specifically, the support information creation unit 3 displays the support information on the screen of the display device constituting the system 1000 or on the screen of the terminal device connected to the system 1000.

When step A3 is executed, the user creates, complements, or modifies the declarative model based on the displayed support information via the declarative model editing component 100. After that, the declarative model is generally used to automatically build (provision) the operating environment according to the configuration. Specifically, a workflow for provisioning is generated according to the definition of a declarative model that has been created, complemented, or corrected, and the system is constructed in the actual environment.

[Effects of embodiments]
As described above, in the present embodiment, the support information for supporting the modification of the declarative model is created. Therefore, in the creation of the declarative model, the declarative model is guided in the direction intended by the user. Information can be presented.

Subsequently, the effect in the present embodiment will be specifically described with reference to FIG. FIG. 6 is a diagram showing the effects obtained by the system construction support device according to the embodiment of the present invention. In the example of FIG. 6, the effect in each of the declarative model creator and the component designer who are users is shown. The effects will be described below for each user.

First, the component creator defines the components (parts) used in the declarative model. Specifically, the component creator provides the infrastructure for operating the business system in the information processing system department of the company. In other words, the component creator will provide the business system designer with the components that can be provided as this basis.

Then, when defining a component, the component creator needs to define without omission what kind of requirements the component has and what ability it can provide. In this case, according to the system construction support device 1 in the present embodiment, the information (dependency information) regarding the dependency of the component can be returned (provided function 1), so that it is "correct" and "use of the local environment". It is possible to create "according to" components.

In addition, the declarative model creator designs a system configuration that is meaningful as a group such as a business system as a declarative model. Specifically, the declarative model creator is the developer of the business system. Traditionally, declarative model creators define declarative models by experience and intuition of dependency definitions between components.

On the other hand, according to the system construction support device 1 in the present embodiment, the support information is provided (provided function 2), so that the declarative model creator is a component when creating the declarative model. A declarative model can be created without defining the dependency definition between them by experience and intuition. The support information provided at this time includes dependency information between the components.

Further, in the present embodiment, since the prediction model is used, it is possible to infer whether or not the workflow operates by the created declarative model without actually provisioning (provided function 2). ).

As described above, according to the present embodiment, the following effects can be obtained in the life cycle of provisioning from the declarative model creation.
Facilitating (efficiency) creation of declarative models.
Automation of dependency completion or correction during editor editing.
Improved accuracy of declarative models (quality of definition).

Further, as described above, in the present embodiment, the accuracy of the declarative model is estimated from the best practices used in the public, the error information in the local environment, etc. while editing the declarative model with the editor. Yes, the guess result can be presented to the user. Further, in the present embodiment, since various information can be collected from the local and the result can be learned, the know-how of the local environment can be encoded.

[program]
The program in the present embodiment may be any program as long as it causes a computer to execute steps A1 to A3 shown in FIG. By installing and executing this program on a computer, the system construction support device and the system construction support method according to the present embodiment can be realized. In this case, the CPU (Central Processing Unit) of the computer functions as an information collection unit 2, a support information creation unit 3, and a prediction model generation unit 4 to perform processing.

Further, the program in the present embodiment may be executed by a computer system constructed by a plurality of computers. In this case, for example, each computer may function as any of the information collection unit 2, the support information creation unit 3, and the prediction model generation unit 4, respectively.

Here, a computer that realizes the system construction support device 1 by executing the program in the present embodiment will be described with reference to FIG. 7. FIG. 7 is a block diagram showing an example of a computer that realizes the system construction support device according to the embodiment of the present invention.

As shown in FIG. 7, the computer 110 includes a CPU 111, a main memory 112, a storage device 113, an input interface 114, a display controller 115, a data reader / writer 116, and a communication interface 117. Each of these parts is connected to each other via a bus 121 so as to be capable of data communication.

The CPU 111 expands the programs (codes) of the present embodiment stored in the storage device 113 into the main memory 112, and executes them in a predetermined order to perform various operations. The main memory 112 is typically a volatile storage device such as a DRAM (Dynamic Random Access Memory). Further, the program in the present embodiment is provided in a state of being stored in a computer-readable recording medium 120. The program in the present embodiment may be distributed on the Internet connected via the communication interface 117.

Further, specific examples of the storage device 113 include a semiconductor storage device such as a flash memory in addition to a hard disk drive. The input interface 114 mediates data transmission between the CPU 111 and an input device 118 such as a keyboard and mouse. The display controller 115 is connected to the display device 119 and controls the display on the display device 119.

The data reader / writer 116 mediates the data transmission between the CPU 111 and the recording medium 120, reads the program from the recording medium 120, and writes the processing result in the computer 110 to the recording medium 120. The communication interface 117 mediates data transmission between the CPU 111 and another computer.

Specific examples of the recording medium 120 include a general-purpose semiconductor storage device such as CF (Compact Flash (registered trademark)) and SD (Secure Digital), a magnetic recording medium such as a flexible disk, or a CD-. Examples include optical recording media such as ROM (Compact Disk Read Only Memory).

The system construction support device 1 in the present embodiment can also be realized by using the hardware corresponding to each part instead of the computer in which the program is installed. Further, the system construction support device 1 may be partially realized by a program and the rest may be realized by hardware.

A part or all of the above-described embodiments can be expressed by the following descriptions (Appendix 1) to (Appendix 12), but the present invention is not limited to the following description.

(Appendix 1)
A device to assist in the construction of a system defined by a declarative model.
An information gathering unit that collects dependency information that identifies the dependencies of the components that make up the declarative model.
Support information that creates support information to support the modification of the declarative model by applying the dependency information to the prediction model that predicts the influence of the dependency of the component on the declarative model. With the creation department
A system construction support device characterized by being equipped with.

(Appendix 2)
It further includes a predictive model generator that uses the dependency information for learning and the declarative model for learning as teacher data, performs machine learning, and generates the predictive model.
The system construction support device according to Appendix 1.

(Appendix 3)
The information gathering unit, as the dependency information,
The information obtained by analyzing the dependency definition of the declarative model that has already been generated,
Information obtained by analyzing errors from the execution results of workflows using the declarative model in the system.
Information obtained by analyzing the definition of the system,
The log held in the system is analyzed, the information obtained by the analysis, and the information regarding the use of the declarative model creation tool published in advance are analyzed, and the information obtained by the analysis is collected. ,
The system construction support device according to Appendix 1 or 2.

(Appendix 4)
The support information creation unit creates, as the support information, information for supporting the generation, complementation, or correction of the dependency between the components in the creation of the declarative model.
The system construction support device according to any one of the appendices 1 to 3.

(Appendix 5)
A way to help build a system defined by a declarative model,
(A) A step that collects dependency information that identifies the dependencies of the components that make up the declarative model.
(B) By applying the dependency information to a prediction model that predicts the influence of the dependency of the component on the declarative model, support information for supporting the modification of the declarative model is created. , Steps and
A system construction support method characterized by having.

(Appendix 6)
(C) Further having a step of performing machine learning to generate the prediction model, using the dependency information for learning and the declarative model for learning as teacher data.
The system construction support method described in Appendix 5.

(Appendix 7)
In the step (a), as the dependency information,
The information obtained by analyzing the dependency definition of the declarative model that has already been generated,
Information obtained by analyzing errors from the execution results of workflows using the declarative model in the system.
Information obtained by analyzing the definition of the system,
The log held in the system is analyzed, the information obtained by the analysis, and the information regarding the use of the declarative model creation tool published in advance are analyzed, and the information obtained by the analysis is collected. ,
The system construction support method according to Appendix 5 or 6.

(Appendix 8)
In the step (c), as the support information, information for supporting the generation, complementation, or correction of the dependency between the components in the creation of the declarative model is created.
The system construction support method according to any one of Supplementary note 5 to 7.

(Appendix 9)
A program to help build a system defined by a declarative model by a computer.
To the computer
(A) A step that collects dependency information that identifies the dependencies of the components that make up the declarative model.
(B) By applying the dependency information to a prediction model that predicts the influence of the dependency of the component on the declarative model, support information for supporting the modification of the declarative model is created. , Steps and
A program that runs.

(Appendix 10)
To the computer
(C) Using the dependency information for learning and the declarative model for learning as teacher data, machine learning is performed to generate the prediction model, and further steps are executed.
The program described in Appendix 9.

(Appendix 11)
In the step (a), as the dependency information,
The information obtained by analyzing the dependency definition of the declarative model that has already been generated,
Information obtained by analyzing errors from the execution results of workflows using the declarative model in the system.
Information obtained by analyzing the definition of the system,
The log held in the system is analyzed, the information obtained by the analysis, and the information regarding the use of the declarative model creation tool published in advance are analyzed, and the information obtained by the analysis is collected. ,
The program according to Appendix 9 or 10.

(Appendix 12)
In the step (c), as the support information, information for supporting the generation, complementation, or correction of the dependency between the components in the creation of the declarative model is created.
The program described in any of Appendix 9-11.

As described above, according to the present invention, in creating a declarative model, it is possible to present information for guiding the declarative model in the direction intended by the user. The present invention is useful for various computer systems in which declarative models are utilized.

1 System construction support device 2 Information collection unit 3 Support information creation unit 4 Prediction model generation unit 5 Prediction model storage unit 6 Dependency bulk data DB
11, 12, 13 Declarative model 21 Workflow 51-55 Analysis result 61 System definition information 62 Usage information 100 Declarative model editing component 101 Model repository 110 Computer 111 CPU
112 Main memory 113 Storage device 114 Input interface 115 Display controller 116 Data reader / writer 117 Communication interface 118 Input device 119 Display device 120 Recording medium 121 Bus 200 Declarative model orchestrator 300 Workflow execution component 510 Declarative model analysis component 520 Workrow execution Situation analysis component 530 Existing system analysis component 540 Log analysis component 550 Public system configuration information Analysis component 400 Execution environment 401 System definition information 402 Log information 600 Internet 601 Public repository group 700 Model component repository 1000 System

Claims (5)

A device to assist in the construction of a system defined by a declarative model.
An information gathering unit that collects dependency information that identifies the dependencies of the components that make up the declarative model.
Support information that creates support information to support the modification of the declarative model by applying the dependency information to the prediction model that predicts the influence of the dependency of the component on the declarative model. With the creation department
A predictive model generation unit that generates the predictive model by performing machine learning using the dependency information for learning and the declarative model for learning as teacher data.
A system construction support device characterized by being equipped with. The information gathering unit, as the dependency information,
The information obtained by analyzing the dependency definition of the declarative model that has already been generated,
Information obtained by analyzing errors from the execution results of workflows using the declarative model in the system.
Information obtained by analyzing the definition of the system,
The log held in the system is analyzed, the information obtained by the analysis, and the information regarding the use of the declarative model creation tool published in advance are analyzed, and the information obtained by the analysis is collected. ,
The system construction support device according to claim 1. The support information creation unit creates, as the support information, information for supporting the generation, complementation, or correction of the dependency between the components in the creation of the declarative model.
The system construction support device according to claim 1 or 2. A way to help build a system defined by a declarative model,
(A) A step that collects dependency information that identifies the dependencies of the components that make up the declarative model.
(B) By applying the dependency information to a prediction model that predicts the influence of the dependency of the component on the declarative model, support information for supporting the modification of the declarative model is created. , Steps and
(C) A step of performing machine learning to generate the prediction model using the dependency information for learning and the declarative model for learning as teacher data.
A system construction support method characterized by having. A program to help build a system defined by a declarative model by a computer.
To the computer
(A) A step that collects dependency information that identifies the dependencies of the components that make up the declarative model.
(B) By applying the dependency information to a prediction model that predicts the influence of the dependency of the component on the declarative model, support information for supporting the modification of the declarative model is created. , Steps and
(C) A step of performing machine learning to generate the prediction model using the dependency information for learning and the declarative model for learning as teacher data.
A program that runs.

Images