Expected configuration generation device, information processing method, program

Abstract

The present invention provides an expected configuration generation device that generates expected configuration information, which shows the components that make a specific design target in a computer system work, as well as the surrounding configurations and the relationships between those components and their surrounding configurations, using a configuration document that describes the components that make the design target work. [Solution] A first prompt information is output to a large-scale language model device, which includes a configuration document describing the components that make a specific design target in a computer system work, and an instruction document that instructs the device to extract components from the configuration document and to extract the peripheral configurations of the components that make the components work properly. Based on the first prompt information, expected configuration information is generated from the basic architecture information, which includes the components and peripheral configurations included in the first analysis result output by the large-scale language model device, describing the components that make the design target work, the peripheral configurations, and the relationship between the components and the peripheral configurations.

Description

Technical Field

[0001] The present disclosure relates to an expected configuration generation device, an information processing method, and a program. Background Art

[0002] Various techniques have been proposed for system design in information and communication technology and the like. For example, Patent Document 1 proposes a method for automatically designing a system configuration that satisfies requirements based on requirement information expressed in graph-like data. Prior Art Documents Patent Documents

[0003] Patent Document 1 International Publication No. 2019 / 216082 Summary of the Invention Problems to be Solved by the Invention

[0004] In the system design apparatus described in Patent Document 1, requirement information of a system required by a user is received as an input, and for each system component (server, NW switch, web application, etc.) included in the requirement information, a design proposal embodying the requirement information is generated based on formalized data of peripheral configurations required for the component to operate normally (hereinafter referred to as "expected configuration").

[0005] In order to realize the operation of such a system design apparatus, it is necessary to grasp in advance information on the expected configuration required for a component to operate, with respect to components that may be included in input requirement information.

[0006] Here, the components include specific components that represent particular products (such as Apache(registered trademark) HTTP Server, Nginx(registered trademark), and IIS (Internet Information Services(registered trademark))) and abstract components that generalize a certain type of product group (such as web servers, approach servers, and database servers). For the latter (abstract components), it is expected that design proposals will be presented with information on general expected configurations that do not presuppose a specific product (for example, products generally classified as web servers operate on an operating system).

[0007] Documents that serve as references when people create expected configurations, such as past system configuration examples and prerequisites for system configurations (e.g., specific middleware is required in the expected configuration for a certain application server to function) (hereinafter referred to as "configuration documents"), may contain descriptions that assume specific products (e.g., Apache® HTTP Server, Nginx®, IIS®).

[0008] The problem with creating an expected configuration using only a specific configuration document as a reference was that the output would include product-specific details based on the assumptions of that document, resulting in an expected configuration that was not representative of a general expected configuration.

[0009] The purpose of this disclosure is to provide an expected configuration generation device, an information processing method, and a program that solve the above-mentioned problems. [Means for solving the problem]

[0010] An expected configuration generation device relating to an illustrative aspect of this disclosure includes: a first prompt information output means that outputs first prompt information to a large-scale language model device, which includes a configuration document describing components that operate a specific design target in a computer system, and an instruction document that instructs the device to extract the components from the configuration document and to extract the peripheral configurations of the components that allow the components to operate normally; and an expected configuration information generation means that generates expected configuration information describing the components that operate the design target, the peripheral configurations, and the relationship between the components and the peripheral configurations from basic architecture information including the components and peripheral configurations included in the first analysis result output by the large-scale language model based on the first prompt information.

[0011] An information processing method relating to an illustrative aspect of this disclosure outputs first prompt information to a large-scale language model device, which includes a configuration document describing the components that make a specific design target in a computer system work, and an instruction document instructing the device to extract the components from the configuration document and to extract the peripheral configurations of the components that make the components work properly. Based on the first prompt information, the device generates expected configuration information which describes the components that make the design target work, the peripheral configurations, and the relationship between the components and the peripheral configurations from basic architecture information which includes the components and peripheral configurations included in the first analysis result output by the large-scale language model.

[0012] A program relating to an illustrative aspect of this disclosure causes the computer of an expected configuration generation device to function as a first prompt information output means that outputs first prompt information to a large-scale language model device, which includes a configuration document describing components that operate a specific design target in a computer system, and an instruction document that instructs the device to extract the components from the configuration document and to extract the peripheral configurations of the components that allow the components to operate normally, and an expected configuration information generation means that generates expected configuration information describing the components that operate the design target, the peripheral configurations, and the relationship between the components and the peripheral configurations from basic architecture information including the components and peripheral configurations included in the first analysis result output by the large-scale language model based on the first prompt information. [Effects of the Invention]

[0013] According to an illustrative aspect of this disclosure, expected configuration information can be generated using a configuration document that describes the components that make a particular design object in a computer system work, showing the components and their surrounding configurations, and the relationships between those components and their surrounding configurations. [Brief explanation of the drawing]

[0014] [Figure 1] This is a block diagram showing the configuration of an expected configuration generating device according to one aspect of the present disclosure. [Figure 2] This figure shows an example of the hardware configuration of an expected configuration generation device according to one aspect of this disclosure. [Figure 3] This is a flowchart illustrating the flow of a model data generation method according to one aspect of this disclosure. [Figure 4] This is a block diagram showing the configuration of an expected configuration generating device according to one aspect of the present disclosure. [Figure 5] The first figure shows an example of a configuration document stored in an expected configuration generation device according to one aspect of this disclosure. [Figure 6] This figure shows an example of data format information stored in an expected configuration generation device according to one aspect of the present disclosure. [Figure 7] It is a diagram illustrating an example of an expected configuration created by an expected configuration generation apparatus according to an aspect of the present disclosure. [Figure 8] It is a diagram illustrating a processing flow of an expected configuration generation apparatus according to an aspect of the present disclosure. [Figure 9] It is a second diagram illustrating an example of a configuration document stored in an expected configuration generation apparatus according to an aspect of the present disclosure. [Figure 10] It is a diagram illustrating a display example according to an aspect of the present disclosure. [Figure 11] It is a functional block diagram of an expected configuration generation apparatus according to another aspect of the present disclosure. [Figure 12] It is a diagram illustrating a processing flow of an expected configuration generation apparatus according to another aspect of the present disclosure. DETAILED DESCRIPTION OF EMBODIMENTS

[0015] Hereinafter, embodiments of the present invention will be exemplified. However, the present invention is not limited to the exemplary embodiments shown below, and various modifications can be made within the scope of the claims. For example, embodiments obtained by appropriately combining techniques (part or all of an article or method) employed in the respective exemplary embodiments shown below are also included within the scope of the present invention. Further, embodiments obtained by appropriately omitting part of the techniques employed in the respective exemplary embodiments shown below are also included within the scope of the present invention. In addition, the effects mentioned in the respective exemplary embodiments shown below are examples of effects expected in the exemplary embodiments, and do not define the scope of the present invention. That is, embodiments that do not achieve the effects mentioned in the respective exemplary embodiments shown below are also included within the scope of the present invention.

[0016] [First Exemplary Embodiment] A first exemplary embodiment, which is an example of an embodiment of the present invention, will be described in detail with reference to the drawings. This exemplary embodiment is a basic form for each of the exemplary embodiments described later. The scope of application of each technology employed in the present exemplary embodiment is not limited to this exemplary embodiment. That is, each technology employed in the present exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure within a range that does not cause any particular technical obstacle. Further, each technology shown in the drawings referenced for describing the present exemplary embodiment can also be employed in other exemplary embodiments included in the present disclosure within a range that does not cause any particular technical obstacle.

[0017] (Configuration of Information Display Device) The expected configuration generation apparatus 10 generates expected configuration information from configuration documents. The configuration documents are documents including past system configuration examples, preconditions for system configuration, configuration procedures, and the like, which serve as reference sources when creating an expected configuration. The expected configuration is data that formalizes peripheral configurations necessary for the elements constituting a system to operate normally.

[0018] The configuration of the expected configuration generation apparatus 10 will be described with reference to Fig. 1. Fig. 1 is a functional block diagram showing the configuration of the expected configuration generation apparatus 10 according to an aspect of the present disclosure. As shown in Fig. 1, the expected configuration generation apparatus 10 includes a basic architecture extraction unit 11, a data format conversion unit 12, and a display unit 13.

[0019] Fig. 2 is a diagram showing an example of a hardware configuration of the expected configuration generation apparatus 10 according to an aspect of the present disclosure. The expected configuration generation device 10 is, for example, a computer equipped with hardware such as a CPU (Central Processing Unit) 101, a GPU (Graphics Processing Unit) 102, a ROM (Read Only Memory) 103, a RAM (Random Access Memory) 104, other storage devices such as an HDD (Hard Disk Drive) 105, and a communication module 106, as shown in Figure 2. The calculation function of the CPU 101 of the expected configuration generation device 10 executes a program, thereby performing functions such as the basic architecture extraction unit 11, the data format conversion unit 12, and the display unit 13.

[0020] The basic architecture extraction unit 11 generates a basic architecture by extracting common information from multiple types of configuration documents. The basic architecture extraction unit 11 can be implemented by any device capable of extracting common information and outputting text, but examples include generative AI (Artificial Intelligence) using a large-scale language model.

[0021] The data format conversion unit 12 converts the basic architecture into the expected configuration format.

[0022] The display unit 13 displays the expected configuration information after conversion. The display unit 13 can be implemented using any device capable of displaying information, such as a liquid crystal display or an organic EL (Electro-Luminescence) display.

[0023] (Information display method flow) The flow of the model data generation method will be explained with reference to Figure 3. Figure 3 is a flowchart showing the flow of the model data generation method. The model data generation method performed by the expected configuration generation device 10 includes a basic architecture extraction process S101, a data format conversion process S102, and a display process S103, as shown in Figure 3.

[0024] In the basic architecture extraction process S101, a basic architecture is generated by extracting common information from multiple types of configuration documents.

[0025] In the data format conversion process S102, the basic architecture is converted to the expected configuration format.

[0026] In display process S103, the expected configuration information after conversion is displayed.

[0027] (Effects of information display methods) As described above, the model data generation method employs a method of extracting information from configuration documents to generate expected configuration information. Therefore, using this model data generation method, users can generate expected configuration information based on configuration documents, without the inclusion of specific product details. This results in the benefit of allowing users to utilize automated design technology without requiring high levels of expertise or high costs.

[0028] [Second exemplary embodiment] A second exemplary embodiment, which is an example of an embodiment of the present invention, will be described in detail with reference to the drawings. Components having the same function as those described in the above-described exemplary embodiment are denoted by the same reference numerals, and their descriptions are omitted as appropriate. The scope of application of each technology adopted in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technology adopted in this exemplary embodiment can also be adopted in other exemplary embodiments included in this disclosure, to the extent that no particular technical problems arise. Furthermore, each technology shown in the drawings referenced to describe this exemplary embodiment can also be adopted in other exemplary embodiments included in this disclosure, to the extent that no particular technical problems arise.

[0029] (Configuration of information display device) The configuration of the expected configuration generation device 10A will be explained with reference to Figure 4. Figure 4 is a functional block diagram showing the configuration of the expected configuration generation device 10A. As shown in Figure 4, in addition to the basic architecture extraction unit 11, data format conversion unit 12, and display unit 13 provided by the expected configuration generation device 10, the expected configuration generation device 10A includes a configuration information database 21 and a data format information database 22.

[0030] Configuration Document 201 contains information describing past system configuration examples, prerequisites, and configuration procedures for system configurations. Examples of the former include the system's basic design document and detailed design document. Examples of the latter include product manuals.

[0031] Figure 5 shows an example of a configuration document 201 according to one aspect of this disclosure. The configuration document 201A shown in Figure 5 is a document describing the operating environment of an application developed by a certain company A. This configuration document 201A indicates that the B operating system is required as the execution environment for the A company application to run. The configuration document 201B shown in Figure 5 is a document describing the operating environment of an application developed by a certain company C. This configuration document 201B indicates that the D operating system or the E operating system is required as the execution environment for the C company application to run.

[0032] The expected configuration generator 10 generates first prompt information, which includes a configuration document describing the components that make a specific design target in a computer system work, and an instruction document that instructs the generator to extract components from the configuration document and to extract the peripheral configurations of the components that make those components work correctly. The expected configuration generator 10 then outputs the first prompt information to the LLM (Large-Scale Language Modeling Device). Based on the first prompt information, the expected configuration generator 10 generates expected configuration information from the basic architecture information, which includes components and peripheral configurations included in the first analysis result output by the LLM, describing common components that make the design target work, the peripheral configurations that make those components work, and the relationship between those components and the peripheral configurations. In this process, the expected configuration generator 10 outputs first prompt information to the Large-Scale Language Modeling Device, which includes configuration documents for multiple products corresponding to a specific design target (application) (for example, specific types of applications from companies A and B), as shown in Figures 5, 201A and 201B, and an instruction document that instructs the generator to extract components from those multiple configuration documents and to extract the peripheral configurations of the components that make those components work correctly.

[0033] Figure 6 shows an example of data format information according to one aspect of this disclosure. The data format information 202A shown in Figure 6 is information that defines a predetermined format and is used to output the components that operate the design target, their peripheral configurations, and the relationship between the components and peripheral configurations in a predetermined format, based on the text containing components and peripheral configurations included in the analysis results output by the large-scale language model. The expected configuration generation device 10A uses this data format information to output expected configuration information that describes the components that operate the design target, their peripheral configurations, and the relationship between the components and peripheral configurations in a predetermined format, based on the text containing components and peripheral configurations included in the analysis results output by the large-scale language model. The expected configuration information is divided into three items: [Name of target component (component)], [Name of peripheral component necessary for the operation of the component (peripheral configuration)], and [Relationship between components (component and peripheral configuration)]. For example, if the target of the design is an application program, and the expected configuration is expressed according to this data format information, such as "the web server is connected to the application server via HTTP / HTTPS communication," then the word "web server" would be entered in the [Name of target component (component)] field, the word "application server" in the [Name of surrounding components required for the operation of the component (peripheral configuration)] field, and the phrase "connected via HTTP / HTTPS communication" in the [Relationship between components (component and peripheral configuration)] field.

[0034] Figure 7 shows an example of expected configuration information according to one aspect of this disclosure. Figure 8 shows the processing flow of an expected configuration generation device according to one aspect of this disclosure. Here, with reference to Figures 4 to 7, an example of the procedure for generating expected configuration information in the expected configuration generation device 10A is shown. First, the basic architecture extraction unit 11 obtains configuration documents 201A (configuration document for a specific type of application a of company A) and 201B (configuration document for a specific type of application b of company C) from the configuration information database 21 (step S101). Note that applications a and b are specific types of applications provided by different companies. For example, Apache® and Nginx®, which are used as web server software, correspond to applications a and b. Configuration document 201A may be a document created by an administrator who built the system in the past, or it may be a document recorded on the internet. Configuration documents 201A and 201B will be collectively referred to as configuration document 201.

[0035] The basic architecture extraction unit 11 generates first prompt information including multiple acquired configuration documents 201 and an instruction document that instructs the extraction of common components from the acquired configuration documents 201 and the extraction of peripheral configurations that enable those components to function correctly (step S102). The instruction document may be a standard document prepared in advance by the administrator of the expected configuration generation device 10A, etc. The basic architecture extraction unit 11 outputs the first prompt information to the LLM (step S103). The LLM outputs a first analysis result, which is the result of inputting the first prompt information. The basic architecture extraction unit 11 acquires the first analysis result (step S104). The basic architecture extraction unit 11 may also perform the functions of the LLM.

[0036] The first analysis result is a document extracted by LLM from multiple configuration documents 201 based on prompt information, describing the general components necessary to operate the system being analyzed, as well as their surrounding configuration. For example, if the system being analyzed is an application, the first analysis result will include a document describing the components necessary to operate that application (web server) and the surrounding configuration necessary to operate those components (application server). This document is called the basic architecture.

[0037] According to the processing of the expected configuration generation device 10A described above, the basic architecture extraction unit 11 can extract from each acquired configuration document what can be considered to be the components of the system to be designed. For example, from configuration document 201A, A company's application (component) and B company's operating system (component) are extracted, and from configuration document 201B, C company's application (component) and D company's operating system (component) are extracted. Based on the analysis results of the LLM, the basic architecture extraction unit 11 can then extract information on the peripheral configurations necessary for each extracted component to function correctly. For example, from configuration document 201A, information such as "A company's application needs to run on B company's operating system to function" is extracted, and from configuration document 201B, information such as "C company's application needs to run on D company's operating system or E company's operating system to function" is extracted.

[0038] Furthermore, according to the processing of the basic architecture extraction unit 11 described above, each extracted component can be classified into a more abstract group based on the LLM processing using the first prompt information. Here, for example, both Company A's application and Company C's application are classified into the more abstract group of "Web server." Similarly, Company B's operating system, Company D's operating system, and Company E's operating system are all classified into the more abstract group of "operating system." Furthermore, according to the processing of the basic architecture extraction unit 11 described above, based on the LLM processing using the first prompt information, information about the peripheral configurations necessary for each component to function correctly can be generated by converting the components included in it into names of more abstract groups. Here, for example, from the information "For Company A's application to function, it must run on Company B's operating system" and the information "For Company C's application to function, it must run on either Company D's or Company E's operating system," LLM generates information (basic architecture information) that is converted into the form "A Web server must run on an operating system." Finally, if similar information is generated as a result of conversion using abstract group names, that information is determined to be basic architecture information. Here, for example, as a result of the conversion using an abstract group name, the information "Web servers need to run on an operating system" appears multiple times, and is therefore determined to be basic architecture information. In the process described above, the basic architecture extraction unit 11 generates prompt information for the LLM to generate such basic architecture information and sends it to the LLM. If the basic architecture extraction unit 11 has the functionality of an LLM, the basic architecture extraction unit 11 may similarly generate basic architecture information.

[0039] The basic architecture extraction unit 11 outputs architecture information, which is a text describing the components to be analyzed and their surrounding configurations included in the first analysis result, to the data format conversion unit 12. The data format conversion unit 12 acquires the architecture information. The data format conversion unit 12 acquires data format information 202A (Figure 6) from the data format information database 22. Based on the data format information, the data format conversion unit 12 generates an expected configuration that describes the components extracted from the text included in the architecture information, the surrounding configurations, and the relationship between the components and the surrounding configurations (step S105).

[0040] More specifically, the data format conversion unit 12 uses the acquired data format information 202A to identify parts of the architecture information, such as constituent elements and peripheral configurations, that need to be replaced through conversion. Here, the data format conversion unit 12 identifies that replacement is necessary for the values ​​of three items: [Name of target component], [Name of peripheral component required for its operation], and [Relationship between components]. The data format conversion unit 12 extracts the information necessary for the above three replacements from the basic architecture acquired from the basic architecture extraction unit 11 and then performs the replacement to generate expected configuration information. For example, from the basic architecture "A web server needs to run on an operating system," the expected configuration information is generated by extracting "Web server" as the value to be entered in the [Name of target component] field, "Operating system" as the value to be entered in the [Name of peripheral component required for its operation] field, and "The former operates on the latter as a logical foundation" as the value to be entered in the [Relationship between components] field.

[0041] The data format conversion unit 12 may also generate expected configuration information using an LLM. For example, the data format conversion unit 12 generates second prompt information including a basic architecture, format definition information, and an instruction document. The instruction information is an instruction document that instructs the LLM to generate expected configuration information that describes, in a predetermined format, the components that operate the design target, the peripheral configuration, and the relationship between the components and the peripheral configuration, from the basic architecture information. The data format conversion unit 12 outputs the second prompt information to the LLM. The LLM generates the expected configuration information using the second prompt information. If the data format conversion unit 12 has the functionality of an LLM, the data format conversion unit 12 may similarly generate the expected configuration information using the LLM.

[0042] The data format conversion unit 12 outputs the generated expected configuration information, along with the basic architecture, to the display unit 13 (step S106).

[0043] The display unit 13 displays the basic architecture and expected configuration (Figure 7) output from the data format conversion unit 12 to a predetermined output destination such as a terminal or display.

[0044] (Effects of information display devices) As described above, the expected configuration generation device 10A employs a configuration that extracts information from the configuration document to generate expected configuration information. Therefore, the expected configuration generation device 10A can generate expected configuration information based on the configuration document, without the inclusion of specific product details. This results in the benefit of allowing users to utilize automated design technology without requiring high levels of expertise or high costs.

[0045] [Third exemplary embodiment] A third exemplary embodiment, which is an example of an embodiment of the present invention, will be described in detail with reference to the drawings. Components having the same function as those described in the above-described exemplary embodiments are denoted by the same reference numerals, and their descriptions are omitted as appropriate. The scope of application of each technology adopted in this exemplary embodiment is not limited to this exemplary embodiment. That is, each technology adopted in this exemplary embodiment can also be adopted in other exemplary embodiments included in this disclosure, to the extent that no particular technical hindrance occurs. Furthermore, each technology shown in the drawings referenced to describe this exemplary embodiment can also be adopted in other exemplary embodiments included in this disclosure, to the extent that no particular technical hindrance occurs.

[0046] Figure 9 is a second figure showing an example of a configuration document according to one aspect of the present disclosure. Configuration document 201C shown in Figure 9 is a document describing the operating environment of an L3SW (Layer 3 switch) developed by a certain company F. This configuration document 201A shows that F's L3SW operates in connection with F's L2SW (Layer 2 switch). Configuration document 201D shown in Figure 9 is a document describing the operating environment of an L3SW developed by a certain company G. This configuration document 201D shows that G's L3SW operates in connection with G's L2SW.

[0047] Next, the processing of the expected configuration generation device in the third exemplary embodiment will be described in the same manner. First, the basic architecture extraction unit 11 retrieves configuration documents 201C and 201D from the configuration information database 21. Configuration documents 201C and 201D may be created by administrators who previously built the system, or they may be information recorded on the internet. Configuration documents 201C and 201B will be collectively referred to as configuration document 201. The basic architecture extraction unit 11 extracts information common to the information contained in the retrieved configuration document 201 as the basic architecture.

[0048] The basic architecture extraction unit 11 generates first prompt information including multiple acquired configuration documents 201 and an instruction document that instructs the extraction of common components from the acquired configuration documents 201 and the extraction of peripheral configurations that enable those components to function correctly (step S102). The instruction document may be a standard document prepared in advance by the administrator of the expected configuration generation device 10A, etc. The basic architecture extraction unit 11 outputs the first prompt information to the LLM (step S103). The LLM outputs a first analysis result, which is the result of inputting the first prompt information. The basic architecture extraction unit 11 acquires the first analysis result (step S104). The basic architecture extraction unit 11 may also perform the functions of the LLM.

[0049] According to the processing performed by the expected configuration generation device 10A described above, the basic architecture extraction unit 11 can extract from each acquired configuration document 201 what can be considered to be components of the system under design. For example, from configuration document 201C, F company's L3SW and F company's L2SW are extracted, and from configuration document 201B, G company's L3SW and G company's L2SW are extracted. Based on the analysis results of the LLM, the basic architecture extraction unit 11 can then extract information on the peripheral configurations necessary for each extracted component to function correctly. For example, from configuration document 201C, information such as "F company's L3SW needs to be connected to F company's L2SW to function" is extracted, and from 201D, information such as "G company's L3SW needs to be connected to G company's L2SW to function" is extracted.

[0050] Furthermore, according to the processing of the basic architecture extraction unit 11 described above, each extracted component can be classified into a more abstract group based on the LLM processing using the first prompt information. Here, for example, both Company F's L3SW and Company G's L3SW are classified into the more abstract group called L3SW. Similarly, both Company F's L2SW and Company G's L2SW are classified into the more abstract group called L2SW. Furthermore, according to the processing of the basic architecture extraction unit 11 described above, based on the LLM processing using the first prompt information, information is generated in which the peripheral configuration information necessary for each component to operate normally is transformed into information in which the components included are named by a more abstract group. Here, for example, from the information "For Company F's L3SW to operate, it needs to be connected to Company F's L2SW" and the information "For Company G's L3SW to operate, it needs to be connected to Company G's L2SW," information is generated that is transformed into the form "For L3SW to operate, it needs to be connected to L2SW." Finally, if similar information is generated as a result of transformation by an abstract group name, that information is determined to be the basic architecture. Here, for example, as a result of the conversion using an abstract group name, the information "L3SW needs to be connected to L2SW to operate" appears multiple times, and this is determined to be the basic architecture. In the process described above, the basic architecture extraction unit 11 generates prompt information for LLM to generate such basic architecture information and sends it to LLM. If the basic architecture extraction unit 11 has the functionality of LLM, the basic architecture extraction unit 11 may similarly generate basic architecture information.

[0051] The basic architecture extraction unit 11 outputs architecture information, which is a text describing the components to be analyzed and their surrounding configurations included in the first analysis result, to the data format conversion unit 12. The data format conversion unit 12 acquires the architecture information. The data format conversion unit 12 acquires data format information 202A (Figure 6) from the data format information database 22. Based on the data format information, the data format conversion unit 12 generates an expected configuration that describes the components extracted from the text included in the architecture information, the surrounding configurations, and the relationship between the components and the surrounding configurations (step S105).

[0052] More specifically, the data format conversion unit 12 uses the acquired data format information 202A to identify parts of the architecture information, such as components and peripheral configurations, that need to be replaced through conversion. Here, the data format conversion unit 12 identifies that replacement is necessary for the values ​​of three items: [name of target component], [name of peripheral component required for its operation], and [relationship between components]. The data format conversion unit 12 extracts the information necessary for the above three replacements from the basic architecture acquired from the basic architecture extraction unit 11 and then performs the replacement to generate the expected configuration information. For example, from the basic architecture "The web server needs to run on the operating system," the expected configuration information is generated by extracting "L3SW" as the value to be entered in the [name of target component] item, "L2SW" as the value to be entered in the [name of peripheral component required for its operation] item, and "they are connected to each other" as the value to be entered in the [relationship between components] item.

[0053] The data format conversion unit 12 may also generate expected configuration information using an LLM. For example, the data format conversion unit 12 generates second prompt information including a basic architecture, format definition information, and an instruction document. The instruction information is an instruction document that instructs the LLM to generate expected configuration information that describes, in a predetermined format, the components that operate the design target, the peripheral configuration, and the relationship between the components and the peripheral configuration, from the basic architecture information. The data format conversion unit 12 outputs the second prompt information to the LLM. The LLM generates the expected configuration information using the second prompt information. If the data format conversion unit 12 has the functionality of an LLM, the data format conversion unit 12 may similarly generate the expected configuration information using the LLM.

[0054] The data format conversion unit 12 outputs the generated expected configuration information, along with the basic architecture, to the display unit 13 (step S106).

[0055] Figure 10 is a second figure showing an example of a representation according to one aspect of the present disclosure. The display unit 13 displays the basic architecture and expected configuration output from the data format conversion unit 12 to a predetermined output destination such as a terminal or display.

[0056] (Effects of information display devices) As described above, the expected configuration generation device 10A employs a configuration that extracts information from the configuration document to generate expected configuration information. Therefore, the expected configuration generation device 10A can generate expected configuration information based on the configuration document, without the inclusion of specific product details. This results in the benefit of allowing users to utilize automated design technology without requiring high levels of expertise or high costs.

[0057] Figure 11 is a functional block diagram of an expected configuration generation device according to another aspect of this disclosure. Figure 12 shows the processing flow of an expected configuration generation device according to another aspect of the present disclosure. The expected configuration generation device 10 performs at least the functions of a prompt information output means 31 and an expected configuration information generation means 32.

[0058] The prompt information output means 31 outputs first prompt information to the large-scale language model device, which includes a configuration document describing the components that make a specific design target in the computer system work, and an instruction document that instructs the device to extract components from the configuration document and to extract the peripheral configurations of the components that make the components work properly (step S301).

[0059] The expected configuration information generation means 32 generates expected configuration information that describes the components that operate the design target, the peripheral configuration, and the relationship between the components and the peripheral configuration, from basic architecture information including components and peripheral configurations included in the first analysis result output by the large-scale language model based on the first prompt information (step S302).

[0060] [Examples of implementation using software] Some or all of the functions of 10, 10A (hereinafter also referred to as "the above devices") may be implemented by hardware such as integrated circuits (IC chips) or by software.

[0061] In the latter case, each of the above devices is implemented, for example, by a computer that executes instructions for a program, which is software that realizes each function. Figure 2 shows an example of such a computer.

[0062] A computer comprises at least one processor and at least one memory. The memory stores programs that cause the computer to operate as each of the above-mentioned devices. In a computer, the processor reads and executes programs from memory, thereby realizing each of the above-mentioned functions.

[0063] For the processor, for example, a CPU (Central Processing Unit), GPU (Graphic Processing Unit), DSP (Digital Signal Processor), MPU (Micro Processing Unit), FPU (Floating Point Number Processing Unit), PPU (Physics Processing Unit), TPU (Tensor Processing Unit), quantum processor, microcontroller, or a combination thereof can be used. For the memory, for example, flash memory, HDD (Hard Disk Drive), SSD (Solid State Drive), or a combination thereof can be used.

[0064] Furthermore, a computer may also be equipped with RAM (Random Access Memory) for deploying programs at runtime and for temporarily storing various data. Additionally, a computer may be equipped with communication interfaces for sending and receiving data with other devices. Furthermore, a computer may be equipped with input / output interfaces for connecting input / output devices such as keyboards, mice, displays, and printers.

[0065] Furthermore, programs can be recorded on tangible, non-temporary recording media that are readable by a computer. Such recording media may include, for example, tape, disk, card, semiconductor memory, or programmable logic circuits. Computers can acquire programs via such recording media. Programs can also be transmitted via transmission media. Such transmission media may include, for example, communication networks or broadcast waves. Computers can also acquire programs via such transmission media.

[0066] Furthermore, each of the above functions of each of the above devices may be implemented by a single processor in a single computer, by multiple processors in a single computer working together, or by multiple processors in each of multiple computers working together. In addition, the programs for implementing each of the above functions in each of the above devices may be stored in a single memory in a single computer, distributed and stored in multiple memories in a single computer, or distributed and stored in multiple memories in each of multiple computers.

[0067] Although the present disclosure has been described above with reference to embodiments, the present disclosure is not limited to the embodiments described above. Various modifications to the structure and details of the present disclosure may be made that can be understood by those skilled in the art within the scope of the present disclosure.

[0068] Some or all of the above embodiments may also be described as follows, but are not limited to the following:

[0069] (Note 1) A first prompt information output means outputs first prompt information to a large-scale language model device, which includes a configuration document describing the components that make a specific design target in a computer system work, and an instruction document that instructs the device to extract the components from the configuration document and to extract the peripheral configurations of the components that make the components work properly. An expected configuration information generation means generates expected configuration information that describes the components that operate the design target, the peripheral configuration, and the relationship between the components and the peripheral configuration, from the basic architecture information including the components and peripheral configuration included in the first analysis result output by the large-scale language model device based on the first prompt information, An expected configuration generation device equipped with the following features.

[0070] (Note 2) The first prompt information output means is The system outputs first prompt information to a large-scale language model device, which includes configuration documents for multiple products corresponding to the specific design target, and an instruction document that instructs the system to extract the components from the configuration documents for the multiple products and to extract the peripheral configurations of the components that enable the components to function correctly. The expected configuration generation device described in Appendix 1.

[0071] (Note 3) The basic architecture information includes the components common to the configuration documents of the multiple products. The expected configuration generation device described in Appendix 2.

[0072] (Note 4) The expected configuration information generation means uses basic architecture information, which includes the constituent elements and peripheral configurations included in the first analysis result output by the large-scale language model device based on the first prompt information, and format definition information that defines a predetermined format for the expected configuration information, to generate expected configuration information from the basic architecture information, which describes the constituent elements that operate the design target, the peripheral configurations, and the relationship between the constituent elements and the peripheral configurations in the predetermined format. An expected configuration generating device as described in any one of the appendices 1 to 3.

[0073] (Note 5) A second prompt information output means outputs second prompt information to a large-scale language model device, which includes the basic architecture information, the formal definition information, and an instruction document that instructs the device to generate expected configuration information that describes the components that operate the design target, the peripheral configuration, and the relationship between the components and the peripheral configuration in a predetermined format, from the basic architecture information. The expected configuration information generation means acquires the expected configuration information included in the second analysis result output by the large-scale language model device based on the second prompt information. The expected configuration generation device described in Appendix 4.

[0074] (Note 6) The design target is a predetermined application program, The aforementioned components are components necessary for the operation of the application program, The aforementioned peripheral configuration is a peripheral configuration necessary for the operation of the aforementioned component. An expected configuration generation device as described in any one of the appendices 1 to 5.

[0075] (Note 7) The design target is a predetermined network device, The aforementioned components are components necessary for the operation of the network device, The aforementioned peripheral configuration is a peripheral configuration necessary for the operation of the aforementioned component. An expected configuration generation device as described in any one of the appendices 1 to 5.

[0076] (Note 8) Output means for outputting the expected configuration information, An expected configuration generating device described in any one of the appendices 1 to 7, comprising the features described.

[0077] (Note 9) First prompt information is output to a large-scale language model device, which includes a configuration document describing the components that make a specific design target in a computer system work, and an instruction document that instructs the device to extract the components from the configuration document and to extract the peripheral configurations of the components that make the components work properly. Based on the first prompt information, the large-scale language model device generates expected configuration information that describes the components that operate the design target, the peripheral configuration, and the relationship between the components and the peripheral configuration from the basic architecture information including the components and peripheral configuration included in the first analysis result output by the large-scale language model device. Information processing methods.

[0078] (Note 10) The system outputs first prompt information to a large-scale language model device, which includes configuration documents for multiple products corresponding to the specific design target, and an instruction document that instructs the system to extract the components from the configuration documents for the multiple products and to extract the peripheral configurations of the components that enable the components to function correctly. The information processing method described in Appendix 9.

[0079] (Note 11) The basic architecture information includes the components common to the configuration documents of the multiple products. The information processing method described in Appendix 10.

[0080] (Note 12) Based on the first prompt information, the large-scale language model device outputs basic architecture information including the components and peripheral configurations included in the first analysis result, and format definition information defining a predetermined format for the expected configuration information, and from the basic architecture information, the device generates expected configuration information that describes the components that operate the design target, the peripheral configurations, and the relationship between the components and the peripheral configurations in the predetermined format. The information processing method described in any one of the appendices 9 through 11.

[0081] (Note 13) Second prompt information, including the basic architecture information, the formal definition information, and an instruction document that instructs the system to generate expected configuration information that describes the components that operate the design target, the peripheral configuration, and the relationship between the components and the peripheral configuration in a predetermined format, is output to the large-scale language model device. Based on the second prompt information, the expected configuration information included in the second analysis result output by the large-scale language model device is acquired. The information processing method described in Appendix 12.

[0082] (Note 14) The design target is a predetermined application program, The aforementioned components are components necessary for the operation of the application program, The aforementioned peripheral configuration is a peripheral configuration necessary for the operation of the aforementioned component. The information processing method described in any one of the appendices 9 through 13.

[0083] (Note 15) The design target is a predetermined network device, The aforementioned components are components necessary for the operation of the network device, The aforementioned peripheral configuration is a peripheral configuration necessary for the operation of the aforementioned component. The information processing method described in any one of the appendices 9 through 13.

[0084] (Note 16) Output the expected configuration information The information processing method described in any one of the appendices 9 through 15.

[0085] (Note 17) The computer of the expected configuration generation device, A first prompt information output means that outputs first prompt information to a large-scale language model device, including a configuration document describing the components that make a specific design target in a computer system work, and an instruction document that instructs the device to extract the components from the configuration document and to extract the peripheral configurations of the components that make the components work properly. An expected configuration information generation means generates expected configuration information that describes the components that operate the design target, the peripheral configuration, and the relationship between the components and the peripheral configuration, from basic architecture information including the components and peripheral configuration included in the first analysis result output by the large-scale language model device based on the first prompt information, A program that makes it function as such.

[0086] (Note 18) The first prompt information output means is The system outputs first prompt information to a large-scale language model device, which includes configuration documents for multiple products corresponding to the specific design target, and an instruction document that instructs the system to extract the components from the configuration documents for the multiple products and to extract the peripheral configurations of the components that enable the components to function correctly. The program described in Appendix 17.

[0087] (Note 19) The basic architecture information includes the components common to the configuration documents of the multiple products. The program described in Appendix 18.

[0088] (Note 20) The expected configuration information generation means uses basic architecture information, which includes the constituent elements and peripheral configurations included in the first analysis result output by the large-scale language model device based on the first prompt information, and format definition information that defines a predetermined format for the expected configuration information, to generate expected configuration information from the basic architecture information, which describes the constituent elements that operate the design target, the peripheral configurations, and the relationship between the constituent elements and the peripheral configurations in the predetermined format. The program described in any one of the appendices 17 through 19.

[0089] (Note 21) The second prompt information output means outputs second prompt information to a large-scale language model device, which includes the basic architecture information, the formal definition information, and an instruction document that instructs the device to generate expected configuration information that describes the components that operate the design target, the peripheral configuration, and the relationship between the components and the peripheral configuration in a predetermined format, based on the basic architecture information. The expected configuration information generation means acquires the expected configuration information included in the second analysis result output by the large-scale language model device based on the second prompt information. The program described in Appendix 20.

[0090] (Note 22) The design target is a predetermined application program, The aforementioned components are components necessary for the operation of the application program, The aforementioned peripheral configuration is a peripheral configuration necessary for the operation of the aforementioned component. The program described in any one of the appendices 17 through 21.

[0091] (Note 23) The design target is a predetermined network device, The aforementioned components are components necessary for the operation of the network device, The aforementioned peripheral configuration is a peripheral configuration necessary for the operation of the aforementioned component. The program described in any one of the appendices 17 through 22.

[0092] (Note 24) This is configured to function as an output means for outputting the aforementioned expected configuration information. The program described in any one of the appendices 17 through 23. [Explanation of symbols]

[0093] 10, 10A...Expected configuration generator 11. Basic Architecture Extraction Unit 12. Data format conversion section 13...Display section 21. Configuration Information Database 22. Data Format Information Database 201...Configuration Document 202...Data Format Information Database

Claims

1. A first prompt information output means outputs first prompt information to a large-scale language model device, which includes a configuration document describing the components that make a specific design target in a computer system work, and an instruction document that instructs the device to extract the components from the configuration document and to extract the peripheral configurations of the components that make the components work properly. An expected configuration information generation means generates expected configuration information that describes the components that operate the design target, the peripheral configuration, and the relationship between the components and the peripheral configuration, from the basic architecture information including the components and peripheral configuration included in the first analysis result output by the large-scale language model device based on the first prompt information, An expected configuration generation device equipped with the following features.

2. The first prompt information output means is The system outputs first prompt information to a large-scale language model device, which includes configuration documents for multiple products corresponding to the specific design target, and an instruction document that instructs the system to extract the components from the configuration documents for the multiple products and to extract the peripheral configurations of the components that enable the components to function correctly. The expected configuration generation apparatus according to claim 1.

3. The basic architecture information includes the components common to the configuration documents of the multiple products. The expected configuration generation apparatus according to claim 2.

4. The expected configuration information generation means uses basic architecture information, which includes the constituent elements and peripheral configurations included in the first analysis result output by the large-scale language model device based on the first prompt information, and format definition information that defines a predetermined format for the expected configuration information, to generate expected configuration information from the basic architecture information, which describes the constituent elements that operate the design target, the peripheral configurations, and the relationship between the constituent elements and the peripheral configurations in the predetermined format. An expected configuration generation apparatus according to any one of claims 1 to 3.

5. A second prompt information output means outputs second prompt information to a large-scale language model device, which includes the basic architecture information, the formal definition information, and an instruction document that instructs the device to generate expected configuration information that describes the components that operate the design target, the peripheral configuration, and the relationship between the components and the peripheral configuration in a predetermined format, from the basic architecture information. The expected configuration information generation means acquires the expected configuration information included in the second analysis result output by the large-scale language model device based on the second prompt information. The expected configuration generation apparatus according to claim 4.

6. The design target is a predetermined application program, The aforementioned components are components necessary for the operation of the application program, The aforementioned peripheral configuration is a peripheral configuration necessary for the operation of the aforementioned component. The expected configuration generation apparatus according to claim 5.

7. The design target is a predetermined network device, The aforementioned components are components necessary for the operation of the network device, The aforementioned peripheral configuration is a peripheral configuration necessary for the operation of the aforementioned component. The expected configuration generation apparatus according to claim 5.

8. Output means for outputting the expected configuration information, An expected configuration generation apparatus according to any one of claims 1 to 3, comprising:

9. First prompt information is output to a large-scale language model device, which includes a configuration document describing the components that make a specific design target in a computer system work, and an instruction document that instructs the device to extract the components from the configuration document and to extract the peripheral configurations of the components that make the components work properly. Based on the first prompt information, the large-scale language model device generates expected configuration information that describes the components that operate the design target, the peripheral configuration, and the relationship between the components and the peripheral configuration from the basic architecture information including the components and peripheral configuration included in the first analysis result output by the large-scale language model device. Information processing methods.

10. The computer of the expected configuration generation device, A first prompt information output means that outputs first prompt information to a large-scale language model device, including a configuration document describing the components that make a specific design target in a computer system work, and an instruction document that instructs the device to extract the components from the configuration document and to extract the peripheral configurations of the components that make the components work properly. An expected configuration information generation means generates expected configuration information that describes the components that operate the design target, the peripheral configuration, and the relationship between the components and the peripheral configuration, from basic architecture information including the components and peripheral configuration included in the first analysis result output by the large-scale language model device based on the first prompt information, A program that makes it function as such.

Images