JP6930141B2 - Monitoring definition generator, monitoring definition generation method, and monitoring definition generation program - Google Patents

Description

The present invention relates to a monitoring definition generator, a monitoring definition generation method, and a monitoring definition generation program.

With the use of cloud computing (hereinafter referred to as cloud) infrastructure and virtualization infrastructure, a system is defined with source code called Infrastructure as Code (IaC), and the system is automatically constructed with a tool called an orchestrator. The way to do it is realized.

A declarative model that is an example of how IaC is defined exists in the format Heat Orchestration Template (HOT) for OpenStack® Heat components. In addition, the declarative model exists in the format of CloudFormation Template (CFN) for AWS Cloudformation (registered trademark) service and in the format of template (template) for Google Cloud Platform (GCP) (registered trademark) Cloud Deployment Manager.

The definition of the declarative model for the orchestra is reused by using a conversion tool or the like. However, the definition of the declarative model is not used as it is. The description of AWS (registered trademark) CFN is compatible with OpenStack Hot.

To address the above issues, for example, the OASIS (Organization for the Advancement of Structured Information Standards) (registered trademark) TOSCA (Topology and Orchestration Specification for), which is a standard specification for defining the system configuration of business systems operating in the cloud. Cloud Applications) makes the definition of system configuration portable by developing standard definitions.

When a declarative model is used, the system department or the like provides a general definition of the declarative model, which is usually called a template. The system builder puts parameters in the provided template and builds the system.

As a technique for improving the accuracy of the template of the declarative model, Patent Document 1 describes a method of improving the template by performing tuning of the declarative model on the execution base of the system and analyzing the construction result of the system. Has been done. That is, Patent Document 1 describes a method for tuning a declarative model that purifies or modifies the declarative model.

Japanese Patent No. 5216853 Japanese Unexamined Patent Publication No. 2016-167179 Japanese Patent No. 4661722

General monitoring settings are incorporated into a general-purpose template created by a highly skilled person in the system department. That is, general monitoring settings are defined in the declarative model.

However, in many cases, the contents of the monitoring settings that meet the non-functional requirements required by the system builder are not incorporated in the template. The declarative model used to build the system is required to reflect information on the non-functional requirements of the system to be built. The non-functional requirements of the system are defined by, for example, SLA (Service Level Agreement).

That is, the system builder is required to tune the declarative model after comprehensively considering the execution environment and the execution status. Also, the definition of monitoring settings depends on the system configuration defined in the declarative model. Therefore, the definition of the monitoring setting is required to be created in consideration of the individual resources of the system.

FIG. 8 is an explanatory diagram showing an example of the relationship between the declarative model, the SLA, and the monitoring setting. Using the declarative model shown in FIG. 8, a system including a load balancer (LB), a Web server (Web), an application server (AP), and a database server (DB) is constructed. As shown in FIG. 8, monitoring settings are required for each communication network device of LB, Web, AP, and DB.

Further, as shown in FIG. 8, auto scaling monitoring settings are required for LB and Web, multiplex processing control monitoring settings for LB and AP, and clustering monitoring settings for DB. In addition, response time monitoring settings are required for the entire system.

Further, as shown in FIG. 8, monitoring settings are required for each of the AP, middleware (MW), patch (Patch), and OS (Operating System) constituting the Web. That is, as shown in FIG. 8, various monitoring settings are required at various locations in order to monitor the SLA.

The tuning method described in Patent Document 1 is not supposed to define monitoring settings. That is, the system builder who uses the tuning method described in Patent Document 1 is required to manually define the monitoring settings. Moreover, it is difficult for the tuning method described in Patent Document 1 to generate a highly accurate definition of a declarative model by the time of the first construction of the system.

Further, Patent Document 2 describes a provisioning support device that automatically supplements a shortage of monitoring settings. However, the provisioning support device described in Patent Document 2 is not a device that generates the contents of the monitoring setting in consideration of the SLA, but is a device that makes up for the shortage of the monitoring setting after executing provisioning. That is, even with the provisioning support device described in Patent Document 2, it is difficult to generate a highly accurate definition of a declarative model by the time of the first construction of the system.

Further, Patent Document 3 describes an operation management system that automatically generates the contents of monitoring settings based on the setting knowledge. However, the operation management system described in Patent Document 3 is a system that automatically generates the content of one monitoring setting that has attracted attention. That is, it is difficult for the operation management system described in Patent Document 3 to automatically generate the contents of the monitoring setting in consideration of the entire system defined by the declarative model.

[Purpose of Invention]
Therefore, the present invention provides a monitoring definition generator, a monitoring definition generation method, and a monitoring definition generation program that can easily generate monitoring definitions in line with the non-functional requirements of a system constructed by a declarative model, which solves the above-mentioned problems. The purpose is to provide.

Monitoring definition generating apparatus according to the present invention, the monitoring definitions associated first reference information indicating a pre-defined criteria for non-functional requirements associated with declarative model and declarative model and declarative model and the first reference information monitoring defined by a template and an input and a second reference information indicating the newly specified criteria for non-functional requirements, it reflects the difference between the first reference information and second reference information to the monitoring definition including bets It is characterized by having a generator that generates a new monitoring definition used in the declarative model based on.

Monitoring definition generating method according to the invention, the monitoring definitions associated first reference information indicating a pre-defined criteria for non-functional requirements associated with declarative model and declarative model and declarative model and the first reference information monitoring defined by a template and an input and a second reference information indicating the newly specified criteria for non-functional requirements, it reflects the difference between the first reference information and second reference information to the monitoring definition including bets It is characterized by generating a new monitoring definition used in the declarative model based on.

The monitoring definition generator according to the present invention relates to a first reference information, a declarative model, and a first reference information that indicate to a computer the pre-defined criteria of the declarative model and the non-functional requirements associated with the declarative model. a template comprising a monitoring definitions, and a second reference information indicating the newly specified criteria for non-functional requirements as an input, to reflect the difference between the first reference information and second reference information to the monitoring definition characterized in that to execute the generation processing for generating a new monitoring definition used by declarative model based on monitoring defined by.

According to the present invention, it is possible to easily generate a monitoring definition according to the non-functional requirements of a system constructed by a declarative model.

It is a block diagram which shows the structural example of the 1st Embodiment of the monitoring definition generation apparatus by this invention. It is a flowchart which shows the operation of the generation processing by the monitoring definition generation apparatus 10 of 1st Embodiment. It is a block diagram which shows the structural example of the 2nd Embodiment of the construction system by this invention. It is explanatory drawing which shows the example of the data model before improvement and the data model after improvement. It is explanatory drawing which shows the use example of the default template generated based on the improved data model. It is an activity diagram which shows the whole operation of the system construction process by the construction system 100 of the 2nd Embodiment. It is an activity diagram which shows the operation of the monitoring definition generation processing by the monitoring definition generation component 131 of the 2nd Embodiment. It is explanatory drawing which shows an example of the relationship between a declarative model, an SLA, and a monitoring setting.

Embodiment 1.
Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram showing a configuration example of a first embodiment of the monitoring definition generator according to the present invention. The monitoring definition generator 10 according to the present invention includes a template including a declarative model and basic information of non-functional requirements related to the declarative model, and basic information of monitoring definitions related to the declarative model and basic information of non-functional requirements. Generation that generates the monitoring definition used in the declarative model by inputting the requirement information indicating a predetermined non-functional requirement and reflecting the difference between the requirement information and the basic information of the non-functional requirement in the basic information of the monitoring definition. A unit 11 (for example, a monitoring definition generation component 131) is provided.

Hereinafter, the generation process by the monitoring definition generation device 10 will be described. FIG. 2 is a flowchart showing the operation of the generation process by the monitoring definition generation device 10 of the first embodiment.

In the generation unit 11, a template including the declarative model and the basic information of the non-functional requirements related to the declarative model and the basic information of the declarative model and the monitoring definition related to the basic information of the non-functional requirements is input (step). S11).

Next, requirement information indicating a predetermined non-functional requirement is input to the generation unit 11 (step S12).

Next, the generation unit 11 generates the monitoring definition used in the declarative model by reflecting the difference between the requirement information and the basic information of the non-functional requirements in the basic information of the monitoring definition (step S13). After the generation, the monitoring definition generation device 10 ends the generation process.

With such a configuration, the monitoring definition generator can easily generate monitoring definitions according to the non-functional requirements of the system constructed by the declarative model.

Further, the monitoring definition generator 10 monitors the system constructed based on the declarative model in which the generated monitoring definition is used by using the monitoring definition (for example, the monitoring setting component 141 and the monitoring component 160). May be provided.

With such a configuration, the monitoring definition generator can monitor whether or not the built system meets the non-functional requirements required by the user.

Further, the monitoring definition generation device 10 may include a construction unit (for example, a declarative model execution component 140) that constructs a system based on a declarative model in which the generated monitoring definition is used.

With such a configuration, the monitoring definition generator can construct a system to be monitored.

The non-functional requirements may also be the requirements defined in the service level aggregate.

With such a configuration, the monitoring definition generator can monitor whether or not the system in which the SLA required by the user is satisfied is satisfied.

Embodiment 2.
[Description of configuration]
Next, a second embodiment of the present invention will be described with reference to the drawings. FIG. 3 is a block diagram showing a configuration example of a second embodiment of the construction system according to the present invention.

The construction system 100 shown in FIG. 3 uses a default template that combines a declarative model, a definition of monitoring settings, and an SLA. In addition, the system parameters input by the user at the time of system construction and the information indicating the SLA required for the system to be constructed are input to the construction system 100.

The construction system 100 shown in FIG. 3 has a function of automatically generating a definition of monitoring settings required when constructing a system based on a declarative model by using a default template and input information.

As shown in FIG. 3, the construction system 100 includes a declarative model editing component 110, a repository 120, a declarative model orchestrator 130, a declarative model execution component 140, an execution environment 150, a monitoring component 160, and the like. Includes service level management component 170.

The declarative model editing component 110 has a function of editing the declarative model. The declarative model editing component 110 outputs the edited declarative model in a source code format such as JSON (JavaScript (registered trademark) Object Notation) or YAML (YAML Ain't Markup Language). The declarative model editing component 110 is, for example, a tool such as an editor having a GUI (Graphical User Interface).

The repository 120 is a database in which the edited declarative model output by the declarative model editing component 110 is stored. The repository 120 is realized by, for example, Subversion (registered trademark) or Git.

The declarative model orchestra 130 has a function of interpreting the declarative model. After interpreting the declarative model, the declarative model orchestrator 130 converts the source code of the declarative model into an internal interpreter so that the construction process is performed. The declarative model orchestrator 130 is realized by, for example, the reference implementation of TOSCA or the Heat implementation of OpenStack.

In the declarative model orchestrator 130 of the present embodiment, a user input parameter, which is a parameter input by the user at the time of system construction, is input. Further, in the declarative model orchestra 130, a definition of a declarative model that can be generally used from the repository 120, that is, a declarative model (template) is input.

Further, as shown in FIG. 3, the declarative model orchestra 130 has a monitoring definition generation component 131. The monitoring definition generation component 131 creates an abstract monitoring definition associated with the declarative model based on the declarative model (template), user input parameters, and information indicating the SLA input to the declarative model orchestrator 130. It has a function to generate.

The monitoring definition generation component 131 of this embodiment is included in the declarative model orchestrator 130, but is a component that is not realized by the TOSCA reference implementation, the OpenStack Heat implementation, or the like. That is, the monitoring definition generation component 131 is a characteristic component of the construction system 100 of the present embodiment.

The declarative model orchestra 130 outputs the definition of the declarative model whose template has been converted so as to be used for system construction. Further, the declarative model orchestrator 130 outputs a monitoring definition that does not depend on the implementation content used for the monitoring setting. The output monitor definition is an abstract monitor definition associated with the output declarative model.

The declarative model execution component 140 has a function of performing a construction process according to the execution environment in which the system is constructed. Specifically, the declarative model execution component 140 performs the system construction process according to the instructions indicated by the declarative model interpreted and transformed by the declarative model orchestra 130.

In order to correspond to various execution environments, the declarative model execution component 140 is generally realized by a plug-in pattern that performs data conversion, API call, etc. corresponding to the execution environment. The function implemented in the declarative model execution component 140 by the plug-in is the function of converting an abstract monitoring definition into a definition that can be interpreted by the corresponding monitoring component 160.

Further, as shown in FIG. 3, the declarative model execution component 140 has a monitoring setting component 141. The monitoring setting component 141 has a function of generating specific monitoring setting parameters corresponding to the monitoring component 160 based on the abstract monitoring definition.

The monitoring setting component 141 inputs the generated monitoring setting parameters to the monitoring component 160. That is, the monitoring setting parameters input by the monitoring setting component 141 to the monitoring component 160 are parameters depending on the implementation contents.

The execution environment 150 is an environment in which the constructed system operates. The execution environment 150 is, for example, a cloud environment.

The monitoring component 160 has a function of monitoring the system operating in the execution environment 150. The monitoring component 160 uses the monitoring setting parameters input from the monitoring setting component 141 for the monitoring process.

The service level management component 170 has a function of managing the service level of the system to be constructed. The service level management component 170 inputs information indicating the SLA required for the system to be constructed into the monitoring definition generation component 131.

FIG. 4 is an explanatory diagram showing an example of a data model before improvement and a data model after improvement. Note that FIG. 4 includes a UML (Unified Modeling Language) class diagram showing definition information of the data model.

FIG. 4A shows a data model before improvement. The “declarative model conforming to a predetermined implementation content” shown in FIG. 4A has a composition relationship with the “monitoring definition (default) depending on the implementation content”. In addition, the "declarative model according to the predetermined implementation content" has no relationship with the "SLA".

That is, as shown in FIG. 4A, the declarative model in the data model before improvement is held in a format including a monitoring definition depending on the implementation content. However, the declarative model in the pre-improvement data model does not retain its relevance to the SLA definition.

FIG. 4B shows the improved data model. The "declarative model" shown in FIG. 4B has an aggregate relationship with the "execution part" and the "variable part". Also, the "declarative model" is related to the "base of SLA definitions". Moreover, the "variable part" depends on the "base of monitoring definition". Also, the "base of monitoring definition" is related to the relationship between the "declarative model" and the "base of SLA definition".

That is, in the improved data model shown in FIG. 4B, the execution part and the variable part of the declarative model are clearly described. In addition, the relationship with SLAs is maintained through the base of monitoring definitions.

Note that FIG. 4C shows a comparison result between the characteristics of the data model before the improvement and the characteristics of the data model after the improvement described above.

FIG. 5 is an explanatory diagram showing an example of using the default template generated based on the improved data model. The default template shown in FIG. 5 corresponds to a declarative model (template) input from the repository 120 shown in FIG. 3 to the declarative model orchestra 130.

FIG. 5 shows an abstract monitoring definition that is output based on a default template that uses the improved data model and information that the user inputs when building the system.

The creator of the default template is the process definition part (execution part shown in FIG. 5), which is the part where the processing method generally provided in the declarative model is described, and the variable definition part, which is the part where the variables are defined. Note that (variable part shown in FIG. 5) is defined separately.

That is, the creator of the default template makes the process definition unit universally reusable. In addition, the creator parameterizes (variables) all the items to which the characteristics of the system to be constructed are input, and while increasing the reusability of the declarative model, the system is constructed according to each system requirement. Create a default template so that.

The creator of the default template is, for example, an operator who has long-term experience in operating the system in the system department or a person in charge of the development department. As mentioned above, the default template reflects the author's knowledge.

Hereinafter, specific examples of the components of the default template shown in FIG. 5, the parameters input at the time of system construction, and the generated monitoring settings will be described.

The "base of SLA definition" is the definition of SLA that is expected to be required for a system built using the declarative model defined in the default template. The SLA is defined as, for example, "transaction processing number: 100 cases / second" and "response time: 5 seconds".

In the "variable part" in the "declarative model", the default values of the parameters used at the time of construction in the system constructed by using the declarative model defined in the default template are defined.

A "monitoring definition base" is a definition of an abstract monitoring setting for a system built using the declarative model defined in the default template.

"SLA" is a definition of SLA required for the system to be constructed. That is, the SLA definition required for systems built using the default template.

"Building parameters" are definitions of building parameters for a system built using the default template. The construction parameters are, for example, various setting items such as machine specification information, minimum and maximum number of machines at the time of autoscale, OS and middleware.

The “monitoring setting” is a definition of an abstract monitoring setting generated by the monitoring definition generation component 131 of the present embodiment.

As mentioned above, the values of the default template reflect, for example, the knowledge of highly skilled personnel in the system department. Further, the value input at the time of system construction reflects, for example, the request of the person in charge of constructing the system.

Multiple SLAs required for a system may include two or more conflicting SLAs in system construction and monitoring. For example, availability and performance are services that are difficult to meet at the same time.

When the above SLA is included, the monitoring definition generation component 131 determines that the relationship defined in the default template is the correct relationship. In addition, the monitoring definition generation component 131 performs correlation analysis and principal component analysis between each value of the default template and each parameter so that the request of the system builder is reflected as much as possible, and then the setting data for monitoring. To generate.

Also, when a system is built using the same versatile default template in different execution environments such as on-premises cloud and public cloud, the request of the system builder should be prioritized as the request for SLA. Sometimes it's good. The monitoring definition generation component 131 may change the monitoring setting data generated according to the situation.

As described above, the construction system 100 of the present embodiment is a system that can automatically generate a monitoring definition corresponding to the declarative model and set a monitoring setting parameter based on the generated monitoring definition in the monitoring component 160. ..

[Explanation of operation]
Hereinafter, the operation of the construction system 100 of the present embodiment will be described with reference to FIGS. 6 to 7.

First, the overall operation of the construction system 100 of the present embodiment will be described with reference to FIG. FIG. 6 is an activity diagram showing the overall operation of the system construction process by the construction system 100 of the second embodiment.

The user builds the system using the default template of the declarative model. Specifically, the user inputs the parameters specified by the user and the information indicating the SLA required for the system to be constructed into the declarative model orchestra 130, and instructs the declarative model orchestra 130 to construct the system. (Step S110).

The user inputs information indicating the SLA to the declarative model orchestra 130 via the service level management component 170.

Next, the declarative model orchestra 130 generates processing information according to the execution environment 150 in which the system is constructed based on the declarative model in the input default template (step S120). The declarative model orchestra 130 inputs the generated processing information to the declarative model execution component 140.

The processing information generated in step S120 is, for example, the source code used to control the construction order. As a specific generation method of processing information, for example, there is a generation method in an implementation example in OpenStack.

Next, the declarative model execution component 140 constructs a system in the execution environment 150 by executing the construction process according to the execution environment 150 using the generated processing information (step S130). The declarative model execution component 140 builds a system using, for example, the external interface of the execution environment 150.

In parallel with the processing of steps S120 to S130, the declarative model orchestrator 130 generates an appropriate monitoring definition for the system to be constructed by using the input default template and parameters (step S140). The monitoring definition generated in step S140 is an abstract monitoring definition.

After the processing of step S130 and the processing of step S140 are completed, the monitoring setting component 141 converts the generated abstract monitoring definition into the monitoring definition for the component to be monitored. The monitoring setting component 141 then sets the converted monitoring definition in the monitoring component 160 (step S150).

Specifically, the monitoring setting component 141 generates monitoring setting parameters based on the abstract monitoring definition. Further, the monitoring setting component 141 inputs the generated monitoring setting parameter to the monitoring component 160.

The monitoring setting component 141 sets the monitoring definition using, for example, the external interface of the monitoring setting component 141. After setting, the construction system 100 ends the system construction process.

Next, the operation of generating the monitoring definition of the declarative model orchestra 130 in step S140 will be described with reference to FIG. 7. FIG. 7 is an activity diagram showing the operation of the monitoring definition generation process by the monitoring definition generation component 131 of the second embodiment.

First, the monitoring definition generation component 131 analyzes the contents of the monitoring settings defined in the default template (that is, the base of the monitoring definition) (step S141). As shown in FIG. 7, the monitoring settings defined in the default template are, for example, operation monitoring, performance monitoring, or log event monitoring.

In addition, the monitoring definition generation component 131 analyzes the difference between the SLA defined in the default template (that is, the base of the SLA definition) and the SLA input at the time of system construction (step S142). For example, the monitoring definition generation component 131 confirms whether or not there is an excess or deficiency between the two.

In addition, the monitoring definition generation component 131 analyzes the declarative model defined in the default template and the parameters input during system construction (step S143).

Next, the monitoring definition generation component 131 receives the analysis results in steps S141 to S143 as inputs, and generates setting data for monitoring to be executed after the system is constructed (step S144).

At the time of generation, the monitoring definition generation component 131 generates monitoring setting data based on the relationship between the value defined in the default template and the value input at the time of system construction, for example.

Next, the monitoring definition generation component 131 creates an abstract monitoring definition based on the monitoring setting data according to the definition of the declarative model so that the generated monitoring setting data is used in the declarative model. Generate (step S145). After generating the abstract monitoring definition, the monitoring definition generation component 131 ends the monitoring definition generation process.

[Explanation of effect]
The construction system of the present embodiment can automatically generate the definition of the monitoring setting according to the non-functional requirements requested by the system builder without depending on the monitoring tool or the operation platform. The reason is that the monitoring definition generation component 131 creates the monitoring setting definition using the default template.

For example, consider the case where the SLA of the system changes. When the number of transactions submitted to the system built with the declarative model is changed from 100 (cases / second) to 50 (cases / second), it is required to define the monitoring settings according to the change of SLA. ..

When the SLA is changed as described above, for example, in the event of a disaster, a production environment system will be installed in a remote data center under the policy of emphasizing business continuity even if processing performance deteriorates. This applies when a similar system (disaster response environment system) is rebuilt.

The monitoring settings for the disaster recovery environment system are the settings in which the normal monitoring settings for various locations shown in FIG. 8 are changed according to the SLA. For example, changes are made to lower the DB response threshold.

The construction system of this embodiment can adjust the monitoring settings according to the non-functional requirements (SLA) of the system constructed by the declarative model by using only the parameters input by the user for system construction. Therefore, the construction system of the present embodiment is preferably used when it is required to create a definition of monitoring settings for the disaster recovery environment system as described above.

The construction system of this embodiment is expected to be suitably used in the fields of cloud management, cloud operation, and system development corresponding to the cloud.

The construction system of the present embodiment is used in combination with the declarative model tuning method described in Patent Document 1, which improves the accuracy of the default template by analyzing the monitoring result after the system is constructed. May be done. The construction system of the present embodiment to which the technique described in Patent Document 1 is applied can generate a more appropriate monitoring definition by executing an improvement loop.

The monitoring definition generation device 10 and the construction system 100 of each embodiment may be realized by, for example, a CPU (Central Processing Unit) that executes processing according to a program stored in a non-temporary storage medium. That is, the generator 11, the declarative model editing component 110, the declarative model orchestrator 130, the declarative model execution component 140, the monitoring component 160, and the service level management component 170 are, for example, by a CPU that executes processing according to program control. It may be realized.

Further, the repository 120 may be realized by, for example, RAM (Random Access Memory).

Further, each part in the monitoring definition generation device 10 and the construction system 100 of each embodiment may be realized by a hardware circuit. As an example, the generator 11, the declarative model editing component 110, the declarative model orchestra 130, the declarative model execution component 140, the monitoring component 160, and the service level management component 170 are each realized by LSI (Large Scale Integration). NS. Further, they may be realized by one LSI.

10 Monitoring definition generation device 11 Generation unit 100 Construction system 110 Declarative model editing component 120 Repository 130 Declarative model orchestrator 131 Monitoring definition generation component 140 Declarative model execution component 141 Monitoring setting component 150 Execution environment 160 Monitoring component 170 Service level management component

Claims (10)

A template comprising a monitoring definitions associated first reference information indicating a pre-defined criteria for non-functional requirements associated with declarative model and the declarative model and said declarative model and the first reference information, the monitoring defined by the as input and the second reference information indicating the newly specified criteria for non-functional requirements, reflects the difference between the second reference information and the first reference information to the monitoring definition A monitoring definition generation device including a generation unit that generates a new monitoring definition used in the declarative model based on the above. The monitoring definition generator according to claim 1, further comprising a monitoring unit that monitors a system constructed based on a declarative model in which the generated new monitoring definition is used, using the new monitoring definition. The monitoring definition generator according to claim 1 or 2, further comprising a construction unit that constructs a system based on a declarative model in which a new generated monitoring definition is used. The monitoring definition generator according to any one of claims 1 to 3, wherein the non-functional requirement is a requirement defined by a service level aggregation. A template comprising a monitoring definitions associated first reference information indicating a pre-defined criteria for non-functional requirements associated with declarative model and the declarative model and said declarative model and the first reference information, the monitoring defined by the as input and the second reference information indicating the newly specified criteria for non-functional requirements, reflects the difference between the second reference information and the first reference information to the monitoring definition A monitoring definition generation method characterized by generating a new monitoring definition used in the declarative model based on the above. The monitoring definition generation method according to claim 5, wherein a system constructed based on a declarative model in which the generated new monitoring definition is used is monitored by using the new monitoring definition. The monitoring definition generator according to claim 5 or 6, wherein the system is constructed based on a declarative model in which the generated new monitoring definition is used. On the computer
A template comprising a monitoring definitions associated first reference information indicating a pre-defined criteria for non-functional requirements associated with declarative model and the declarative model and said declarative model and the first reference information, the monitoring defined by the as input and the second reference information indicating the newly specified criteria for non-functional requirements, reflects the difference between the second reference information and the first reference information to the monitoring definition A monitoring definition generation program for executing a generation process for generating a new monitoring definition used in the declarative model based on the above. On the computer
The monitoring definition generation program according to claim 8, wherein a monitoring process for monitoring a system constructed based on a declarative model in which the generated new monitoring definition is used is executed using the new monitoring definition. On the computer
The monitoring definition generation program according to claim 8 or 9, wherein a construction process for constructing a system based on a declarative model in which a new generated monitoring definition is used is executed.

Images