Program and design support system

Abstract

This program and design support system provide the ability to quickly propose cloud architectures, which are design information for combinations of multiple cloud computing services, to users. [Solution] Program 12a causes the processor of the design support system 100 to execute the following processes: acquiring requirement information in natural language input by the user from the user terminal 20; generating a prompt based on the requirement information that includes an instruction to the artificial intelligence system 30 to output the necessary cloud computing services; inputting the prompt to the artificial intelligence system 30 and acquiring the output result from the artificial intelligence system 30; generating proposed cloud architecture information based on the output result; and sending the proposed cloud architecture information to the user terminal 20.

Description

【Technical Field】 【0001】 The present invention relates to a program and a design support system. 【Background Art】 【0002】 Patent Document 1 discloses a proposal system that proposes services realized on the cloud system side to users. This proposal system selects, based on the characteristics of an image captured by a user with a camera, a service to be proposed to the user among the services related to image analysis realized on the cloud system side. Then, this proposal system causes the terminal device to execute a process of proposing the selected service to the user. 【Prior Art Documents】 【Patent Documents】 【0003】 【Patent Document 1】 Japanese Patent Application Laid-Open No. 2024-50036 【Summary of the Invention】 【Problems to be Solved by the Invention】 【0004】 Here, there are a large number (at least 100 or more) of cloud computing services. A user needs to consider which combination (or design) should be made from a large number of cloud computing services for the specifications they desire. Also, even a proposer who proposes a combination of cloud computing services to the user needs a lot of consideration time regarding which combination (or design) should be made from a large number of cloud computing services. However, although the proposal system described in Patent Document 1 can propose services related to image analysis realized on the cloud system side, it cannot propose a combination of multiple cloud computing services. 【0005】 This invention was made to solve the above-mentioned problems, and aims to provide a program and design support system that can quickly propose a cloud architecture, which is design information for a combination of multiple cloud computing services, to a user. [Means for solving the problem] 【0006】 To achieve the above objective, the program disclosed below is a program that causes the processor of a design support system to execute control processing. The program causes the processor to execute the following: a process of acquiring requirements information in natural language input by the user from a user terminal; a process of generating a prompt based on the requirements information, which includes an instruction to output the necessary cloud computing services to the artificial intelligence system; a process of inputting the prompt to the artificial intelligence system and acquiring the output result from the artificial intelligence system; a process of generating proposed cloud architecture information, which is design information of a combination of multiple cloud computing services, based on the output result; and a process of sending the proposed cloud architecture information to the user terminal. 【0007】 Furthermore, the design support system disclosed below includes: a requirements information acquisition unit that acquires requirements information in natural language input by the user from a user terminal; a prompt generation unit that generates a prompt including an instruction to the artificial intelligence system to output the necessary cloud computing services based on the requirements information; an output result acquisition unit that inputs the prompt to the artificial intelligence system and acquires the output result from the artificial intelligence system; a proposal information generation unit that generates proposed cloud architecture information, which is design information of a combination of multiple cloud computing services, based on the output result; and a transmission unit that transmits the proposed cloud architecture information to the user terminal. [Effects of the Invention] 【0008】 With the above configuration, it is possible to quickly propose a cloud architecture, which is design information for a combination of multiple cloud computing services, to the user. [Brief explanation of the drawing] 【0009】 [Figure 1] Figure 1 is a schematic diagram showing the configuration of the design support system 100 in this embodiment. [Figure 2] Figure 2 is a block diagram showing the configuration of the design support system 100 according to this embodiment. [Figure 3] Figure 3 is a block diagram showing the configuration of the user terminal 20. [Figure 4] Figure 4 is a diagram illustrating the input and output of the artificial intelligence system 30. [Figure 5] Figure 5 is a functional block diagram of the artificial intelligence control unit 11a. [Figure 6] Figure 6 shows an example of a selection screen displayed on the display unit 23 of the user terminal 20. [Figure 7] Figure 7 shows an example of a requirements input screen displayed on the display unit 23 of the user terminal 20. [Figure 8] Figure 8 shows an example of a prompt generated by the artificial intelligence control unit 11a. [Figure 9] Figure 9 shows an example of a suggestion screen displayed on the display unit 23 of the user terminal 20. [Figure 10] Figure 10 shows an example of a feedback input screen displayed on the display unit 23 of the user terminal 20. [Figure 11] Figure 11 shows an example of a prompt that the lead agent 11b sends to the configuration diagram / infrastructure configuration file generation agent 11cb. [Figure 12] Figure 12 shows an example of an infrastructure configuration file displayed on the display unit 23 of the user terminal 20. [Figure 13]FIG. 13 is a diagram showing an example of a prompt that the lead agent 11b transmits to the configuration diagram / infrastructure configuration file generation agent 11cb. [Figure 14] FIG. 14 is a diagram showing an example of what is displayed on the display unit 23 of the user terminal 20. [Figure 15] FIG. 15 is a diagram showing an example of a prompt that the lead agent 11b transmits to the evaluation agent 11cd. [Figure 16] FIG. 16 is a diagram showing an example of an evaluation result displayed on the display unit 23 of the user terminal 20. [Figure 17] FIG. 17 is a diagram showing an example of a screen displaying the cost estimate displayed on the display unit 23 of the user terminal 20. [Figure 18] FIG. 18 is a diagram showing an example of a configuration diagram generation input screen displayed on the display unit 23 of the user terminal 20. [Figure 19] FIG. 19 is a diagram showing an example of a cost estimate input screen displayed on the display unit 23 of the user terminal 20. [Figure 20] FIG. 20 is a diagram showing an example of an evaluation generation input screen displayed on the display unit 23 of the user terminal 20. 【Embodiments for Carrying Out the Invention】 【0010】 Hereinafter, an embodiment of the present invention will be described based on the drawings. Note that the present invention is not limited to the following embodiments, and design changes can be made as appropriate within the scope that satisfies the configuration of the present invention. Also, in the following description, the same parts or parts having the same function are commonly used with the same reference numerals among different drawings, and the repeated description thereof is omitted. In addition, the respective configurations described in the embodiments and modification examples may be combined or changed as appropriate. Also, in order to make the description easier to understand, in the drawings referred to below, the configuration is shown in a simplified or schematic manner, or some constituent members are omitted. 【0011】 [Overall Configuration of the Design Support System 100] FIG. 1 is a schematic diagram showing the configuration of the design support system 100 in the present embodiment. The design support system 100 according to the present embodiment is a system that proposes a cloud architecture, which is design information of a combination of a plurality of cloud computing services. The "cloud computing service" is a service in which software programs and data are stored on the cloud and can be used by a user accessing the cloud. Further, the design support system 100 is a system that executes control processing so as to function as a plurality of artificial intelligence agents that execute input of instructions to the artificial intelligence system 30 and acquisition of outputs from the artificial intelligence system 30 on behalf of the user. The "artificial intelligence agent" is a software program that provides an answer to the user so as to satisfy the user's request by performing input / output on behalf of the user to the artificial intelligence system 30. In the present disclosure, the "system" can adopt any known hardware configuration (control circuit, memory circuit, network, etc.) other than those exemplified below. 【0012】 As shown in FIG. 1, the artificial intelligence control unit 11a in the design support system 100 functions as a plurality of artificial intelligence agents. The plurality of artificial intelligence agents include a lead agent 11b and a plurality of sub-agents 11c. The design support system 100 according to the present embodiment includes a lead agent 11b specialized in analyzing requirement information input by the user and a plurality of sub-agents 11c specialized in specific functions other than the analysis of requirement information, which execute the processing for the artificial intelligence system 30 separately. In other words, in the present embodiment, the plurality of artificial intelligence agents are configured in multiple layers. Thereby, even when there are constraints on the context in the artificial intelligence system 30, it is possible to accurately obtain an output from the artificial intelligence system 30. 【0013】 In this embodiment, the user inputs requirements information into the user terminal 20 in natural language. The design support system 100 acquires the requirements information and generates a prompt based on the requirements information, which includes an instruction to the artificial intelligence system 30 to output the necessary cloud computing services. The design support system 100 then inputs the prompt into the artificial intelligence system 30 and acquires the output result from the artificial intelligence system 30. The design support system 100 generates proposed cloud architecture information based on the output result. The design support system 100 then sends the proposed cloud architecture information to the user terminal 20. 【0014】 As a result, when a user inputs requirements information in natural language into the user terminal 20, proposed cloud architecture information is sent to the user terminal 20. Consequently, users do not need to consider (design) combinations of multiple cloud computing services from a large number of cloud computing services, and can quickly obtain proposed cloud architectures. 【0015】 [Configuration of each part of the design support system 100] Figure 2 is a block diagram showing the configuration of the design support system 100 according to this embodiment. As shown in Figure 2, the design support system 100 is configured to communicate with multiple user terminals 20 and an artificial intelligence system 30 via a network N. The network N is, for example, a Local Area Network (LAN) and the Internet, but other networks may also be used. 【0016】 As shown in Figure 1, the design support system 100 includes a server device 10. The server device 10 includes a control unit 11, a storage unit 12, and a communication unit 13. The server device 10 may be an on-premise (stationary) server device or a server device configured on the cloud. The control unit 11 includes a processor that executes control processing by executing program 12a. The control unit 11 then executes each control processing of the server device 10. The processor executes control processing to propose a cloud architecture, which is design information for a combination of multiple cloud computing services. The processor also executes control processing to function as multiple artificial intelligence agents that, on behalf of the user, input instructions to the artificial intelligence system 30 and acquire outputs from the artificial intelligence system 30. Program 12a is a software program configured to make the processor function as multiple artificial intelligence agents. The storage unit 12 includes ROM (Read Only Memory) and RAM (Random Access Memory). The storage unit 12 stores program 12a, an output database 12b, and a history record database 12c. The deliverables database 12b stores data on design support know-how accumulated by the entity operating the design support system 100, separate from the data learned by the artificial intelligence system 30 and the data collected by the artificial intelligence system 30. The history record database 12c stores a history of cloud architecture proposal information that has been proposed to users to date. When the artificial intelligence agent of the design support system 100 performs its functions, it can refer to the deliverables database 12b and the history record database 12c to output proposal information with higher accuracy. The communication unit 13 includes a communication interface connected to the network N. 【0017】 (Configuration of user terminal 20) Figure 3 is a block diagram showing the configuration of the user terminal 20. The user terminal 20 is an information processing terminal used by the user (e.g., a personal computer, tablet, smartphone, etc.). The user terminal 20 includes a control unit 21, an operation unit 22, a display unit 23, a communication unit 24, a microphone 25, a speaker 26, and a storage unit 27. The control unit 21 includes a processor that executes control processing by running a program. The operation unit 22 is, for example, a keyboard, mouse, and touch panel. The display unit 23 is, for example, an organic EL display or a liquid crystal display. The communication unit 24 is a communication interface for communicating with the network N. The microphone 25 converts the user's voice into audio data (electrical signals). The speaker 26 outputs audio data acquired from the design support system 100 via the communication unit 24. The storage unit 27 includes ROM (Read Only Memory) and RAM (Random Access Memory). 【0018】 (Configuration of artificial intelligence system 30) Figure 4 is a diagram illustrating the input and output of the artificial intelligence system 30. The artificial intelligence system 30 is, for example, a generative artificial intelligence (AI) system. As shown in Figure 2, the artificial intelligence system 30 includes a pre-trained large-scale language model (LLM) 41. When a prompt is input, the artificial intelligence system 30 generates an output result based on the large-scale language model (LLM) 41. A "prompt" is an instruction or question that commands the artificial intelligence system 30 to generate an output result. The prompt can be, for example, natural language, but may also be a formal language (e.g., computer language) depending on the artificial intelligence system 30. 【0019】 (Configuration of the control unit 11) As shown in Figure 1, the control unit 11 includes an artificial intelligence control unit 11a and a memory control unit 11d. The memory control unit 11d refers to the deliverable database 12b and the history record database 12c when the artificial intelligence control unit 11a generates proposed information for the cloud architecture. The memory control unit 11d stores the generated proposed information for the cloud architecture in the history record database 12c. 【0020】 Figure 5 is a functional block diagram of the artificial intelligence control unit 11a. As shown in Figure 5, the artificial intelligence control unit 11a functions as a requirements information acquisition unit 11aa, a prompt generation unit 11ab, an output result acquisition unit 11ac, a proposal information generation unit 11ad, and a transmission control unit 11ae by executing program 12a. 【0021】 (Control processing by artificial intelligence agents) As shown in Figure 1, the artificial intelligence control unit 11a functions as a lead agent 11b and multiple sub-agents 11c for the artificial intelligence system 30. The lead agent 11b functions as a requirements information acquisition unit 11aa that acquires requirements information in natural language input by the user from the user terminal 20. The lead agent 11b also functions as a transmission control unit 11ae that transmits proposed cloud architecture information to the user terminal 20. The lead agent 11b and the multiple sub-agents 11c function as a prompt generation unit 11ab, an output result acquisition unit 11ac, and a proposed information generation unit 11ad, respectively. 【0022】 Figure 6 shows an example of a selection screen displayed on the display unit 23 of the user terminal 20. The artificial intelligence control unit 11a displays a selection screen on the display unit 23 of the user terminal 20, allowing the user to select one of the following: an image instructing the generation of proposed cloud architecture information (in the example of Figure 6, "Design Generation Flow (Guided Process)"), an image instructing the generation of infrastructure configuration files and configuration diagrams, an image instructing the estimation of costs, and an image instructing evaluation. When the user selects the image instructing the generation of proposed cloud architecture information, the artificial intelligence control unit 11a transitions the screen displayed on the display unit 23 from the selection screen to the requirements input screen shown in Figure 7. When the user selects the image instructing the generation of infrastructure configuration files and configuration diagrams, the artificial intelligence control unit 11a transitions the screen displayed on the display unit 23 from the selection screen to the configuration diagram generation input screen shown in Figure 12. When the user selects the image instructing the estimation of costs, the artificial intelligence control unit 11a transitions the screen displayed on the display unit 23 from the selection screen to the cost estimation input screen shown in Figure 13. When the user selects an image to be evaluated, the artificial intelligence control unit 11a transitions the screen displayed on the display unit 23 from the selection screen to the evaluation generation input screen shown in Figure 14. 【0023】 Figure 7 shows an example of a requirements input screen displayed on the display unit 23 of the user terminal 20. As shown in Figure 7, the requirements information acquisition unit 11aa displays the requirements input screen on the display unit 23 of the user terminal 20. The requirements input screen includes input fields for overview and purpose, budget, and performance requirements. The overview and purpose input field accepts, for example, "system requirements" (an overview of the system desired by the user and the purpose of the system) in natural language. In the example in Figure 7, the following is entered into the input field: "I want to build a new e-commerce site. I anticipate 100,000 page views per month, and since access may surge during sales, I would like it to be able to automatically scale up and down. User information and product information will be managed in a database. I want static content such as images to be delivered at high speed. Security should comply with basic best practices." 【0024】 In the budget input field, for example, a monthly budget estimate (500 (dollars) in the example in Figure 7) is entered. In the performance requirements input field, for example, a response time (within 200ms in the example in Figure 7) is entered. The requirements input screen also includes an input field for the user's preference for using a specific cloud computing service. If the user has already identified the cloud computing service to use, the name of the cloud computing service is entered in the input field for the user's preference for using a specific cloud computing service. The requirements input screen displays a "Generate Design Proposal" button, and when the user selects this button, the requirements information acquisition unit 11aa acquires the information entered in each input field as requirements information. 【0025】 Figure 8 shows an example of a prompt generated by the artificial intelligence control unit 11a. As shown in Figure 8, the prompt generation unit 11ab generates prompts based on requirement information. The prompt includes instructions to output the necessary cloud computing services to the artificial intelligence system. The prompt also includes instructions to analyze the requirement information (system requirements). For example, in the prompt, the requirement information obtained from the user terminal 20 is described in the "User Request" and "Requirements" fields. The prompt also includes instructions such as, "Analyze the following system requirements and propose the optimal architecture." 【0026】 As shown in Figure 8, the prompt includes instructions on the perspectives for analysis. For example, the instructions on the perspectives for analysis include instructions to analyze from the following perspectives: required cloud computing services, architectural patterns which represent the structure of the required cloud computing services, security requirements (security considerations), scalability requirements, and cost optimization. This allows the artificial intelligence system 30 to perform analyses not only on the required cloud computing services but also on architectural patterns, security requirements, and cost optimization, and to provide the user with suggested information based on these analyses. 【0027】 Furthermore, prompts include instructions for the processing steps. For example, a prompt may include instructions to process in the following order: analyze the request information, collect information, infer and integrate, and then structure and respond. An instruction for "analyze the request information" might be, for example, "Analyze the user's request in detail and identify the necessary information." An instruction for "collect information" might be, for example, "Use available tools to collect official information." The artificial intelligence system 30 is configured to access the systems (servers) of cloud computing service providers via the network N and obtain official information. This allows the artificial intelligence system 30 to collect information on multiple cloud computing services and then propose the necessary cloud computing services, even if it has not yet learned information on multiple cloud computing services. 【0028】 Instructions to "infer and integrate" are, for example, instructions such as "Infer the optimal solution based on the collected information." Instructions to "structure and respond" are instructions that specify the response format, for example, instructions such as "Generate a response in the following format." Instructions that specify the response format include providing an output field for a summary of the user's request information, an output field for a recommended solution (proposal information for a specific cloud architecture), an output field for technical details (details necessary for implementation), an output field for a cost estimate, an output field for the next steps (specific actions toward implementation), and an output field for additional context. The artificial intelligence control unit 11a (output result acquisition unit 11ac) inputs the generated prompt to the artificial intelligence system 30 and acquires the output result from the artificial intelligence system 30. 【0029】 As shown in Figure 1, the multiple sub-agents 11c include an architect design agent 11ca, a configuration diagram / infrastructure configuration file generation agent 11cb, a cost estimation agent 11cc, and an evaluation agent 11cd. The prompt generation unit 11ab (lead agent 11b) sends the above prompt to the architect design agent 11ca. The architect design agent 11ca (output result acquisition unit 11ac) inputs the prompt to the artificial intelligence system 30 and acquires the output result from the artificial intelligence system 30. 【0030】 Figure 9 shows an example of a proposal screen displayed on the display unit 23 of the user terminal 20. The lead agent 11b obtains output results from the architect design agent 11ca. Based on the output results, the lead agent 11b (proposal information generation unit 11ad) generates proposal information for the cloud architecture and displays it as a proposal screen on the display unit 23 of the user terminal 20. The proposal screen displays an "overview," service configuration, and features as the analysis results of the requirements information. Since the analysis results of the requirements information are displayed, the user can check whether the analysis results of the requirements information match the requirements they have in mind. The overview includes the cloud computing service name and function corresponding to the analysis results. The service configuration associates the necessary cloud computing service name and function. The features include the features and description of the proposed cloud architecture. As described above, if the user inputs requirements information in natural language into the user terminal 20, the proposal information for the cloud architecture is sent to the user terminal 20. This eliminates the need for the user to consider (design) combinations of multiple cloud computing services from a large number of cloud computing services, and the user can quickly obtain a cloud architecture proposal. 【0031】 Figure 10 shows an example of a feedback input screen displayed on the display unit 23 of the user terminal 20. As shown in Figure 9, the proposal screen displays two options: "Approve and proceed" to approve the proposed cloud architecture, and "Reject and provide feedback" to reject the proposed cloud architecture. If "Approve and proceed" is selected, the prompt generation unit 11ab (lead agent 11b) generates a further prompt based on the output from the artificial intelligence system 30 and sends it to one of the multiple sub-agents 11c. If "Reject and provide feedback" is selected, as shown in Figure 10, the lead agent 11b displays a feedback input screen on the display unit 23 of the user terminal 20. The feedback input screen displays a feedback input field that accepts input from the user in natural language, and an image for selecting "Regenerate design proposal". The requirements information acquisition unit 11aa acquires the information entered in the feedback input field. The prompt generation unit 11ab generates a prompt to give a correction instruction based on the feedback information and sends the prompt to the architect design agent 11ca. The architect design agent 11ca inputs prompts to the artificial intelligence system 30, retrieves output results from the artificial intelligence system 30, and sends the output results to the lead agent 11b. The lead agent 11b then displays the proposed revision information for the cloud architecture (revised proposal screen) on the display unit 23 of the user terminal 20. This process of feedback is repeated until the user is satisfied, and the proposed cloud architecture is revised. 【0032】 Figure 11 shows an example of a prompt that the lead agent 11b sends to the configuration diagram / infrastructure configuration file generation agent 11cb. As shown in Figure 11, the prompt generation unit 11ab (lead agent 11b) generates a prompt that includes instructions to generate an infrastructure configuration file (a blueprint defined in code) based on the output results from the artificial intelligence system 30. The lead agent 11b sends the prompt shown in Figure 11 to the configuration diagram / infrastructure configuration file generation agent 11cb. For example, the prompt includes a role specification (in the example in Figure 11, "You are an expert in XX infrastructure configuration"), instructions to generate an infrastructure configuration file based on the output results (architecture information), instructions to specify the perspective to be used during generation, and instructions to specify the output format. Note that "○", "△", and "□" are redacted to indicate service names, product names, etc., in order to explain this embodiment. 【0033】 The "perspectives for generation" are the items to be evaluated, such as "best practices of the ○○ evaluation framework," "security settings (principle of least privilege)," "high availability and scalability," "cost efficiency," "operability and maintainability," "dependencies between resources," and "reusability through parameterization." The prompts also include instructions for items to be included in the infrastructure configuration file (e.g., "appropriate parameter definitions," "lowest privilege settings for security groups," "appropriate settings for □□ roles and policies," "resource management through tagging," and "output definitions"). The prompts also include instructions to specify the format and instructions to specify security best practices (e.g., "buckets must be configured with ×××," "△△ must have encryption enabled," "□□ roles must adhere to the principle of least privilege," "security groups must open the minimum necessary ports," and "all resources must be appropriately tagged"). 【0034】 Figure 12 shows an example of an infrastructure configuration file displayed on the display unit 23 of the user terminal 20. The infrastructure configuration file generation agent 11cb inputs a prompt to the artificial intelligence system 30, obtains the output result from the artificial intelligence system 30, and sends the output result to the lead agent 11b. Based on the output result, the lead agent 11b displays an image showing the infrastructure configuration file shown in Figure 12 as a suggestion screen on the display unit 23 of the user terminal 20. The suggestion screen displays the option "Approve and proceed" to approve the suggested infrastructure configuration file and the option "Deny and provide feedback" to reject the suggested infrastructure configuration file. If "Deny and provide feedback" is selected, the feedback input screen shown in Figure 10 is displayed on the display unit 23 of the user terminal 20. 【0035】 Figure 13 shows an example of a prompt that the lead agent 11b sends to the configuration diagram / infrastructure configuration file generation agent 11cb. As shown in Figure 13, the prompt generation unit 11ab (lead agent 11b) generates prompts that include instructions to generate an infrastructure configuration file and a configuration diagram based on the output results from the artificial intelligence system 30. The prompt instructing the generation of the infrastructure configuration file is the same as described above and is therefore omitted. The lead agent 11b sends the prompt shown in Figure 13 to the configuration diagram / infrastructure configuration file generation agent 11cb. For example, the prompt includes a role specification (in the example in Figure 13, "You are an expert in XX infrastructure configuration and infrastructure diagrams"), an instruction to generate a configuration diagram based on the output results (architecture information), an instruction to specify the output format "XX diagram-as-code format", and an instruction to specify things to consider during generation (for example, "Compliance with XX best practices", "Appropriate implementation of security settings", "Accurate definition of dependencies between resources", and "A configuration that is actually deployable"). The prompt contains instructions to make the output usable for deployment (placing the developed application or system in an actual operating environment and running it), rather than just plain text. Although not shown in Figure 13, the prompt may also include instructions to specify the output format of the configuration diagram (e.g., PNG, JPG, or SVG format). 【0036】 Figure 14 shows an example of what is displayed on the display unit 23 of the user terminal 20. The configuration diagram / infrastructure configuration file generation agent 11cb inputs a prompt to the artificial intelligence system 30, obtains the output result from the artificial intelligence system 30, and sends the output result to the lead agent 11b. Based on the output result, the lead agent 11b displays the configuration diagram shown in Figure 14 as a proposal screen on the display unit 23 of the user terminal 20. The proposal screen displays the option to approve the proposed configuration diagram, "Approve and proceed to the next step," and the option to reject the proposed configuration diagram, "Reject and provide feedback." If "Reject and provide feedback" is selected, the feedback input screen shown in Figure 10 is displayed on the display unit 23 of the user terminal 20. 【0037】 Figure 15 shows an example of a prompt that lead agent 11b sends to evaluation agent 11cd. As shown in Figure 15, the prompt generation unit 11ab (lead agent 11b) generates a prompt that includes instructions to evaluate proposed cloud architecture information (infrastructure configuration file). The prompt is an evaluation instruction prompt that includes instructions to evaluate the proposed cloud architecture information (infrastructure configuration file) from the following evaluation perspectives (evaluation items): operational excellence, security, reliability, performance efficiency, and cost optimization. The prompt also includes instructions to provide answers regarding "good points," "concerns," "suggestions for improvement," and "specific implementation methods" in the evaluation. The above evaluation perspectives (evaluation items) may be, for example, evaluation perspectives (evaluation items) specified by a company providing a certain cloud computing service. 【0038】 Figure 16 shows an example of evaluation results displayed on the display unit 23 of the user terminal 20. The evaluation agent 11cd inputs a prompt to the artificial intelligence system 30, obtains output results from the artificial intelligence system 30, and sends the output results to the lead agent 11b. Based on the output results, the lead agent 11b displays the evaluation results for cost optimization, operational excellence, security, reliability, and performance efficiency on the display unit 23 of the user terminal 20. This allows the user to check the evaluation results. As shown in Figure 16, the lead agent 11b displays a score for each evaluation item (evaluation perspective) and displays the average value (or sum) of all evaluation items as an overall score on the display unit 23. The evaluation results include strengths, concerns, and improvement suggestions for each evaluation item. 【0039】 Figure 17 shows an example of a screen displaying an estimated cost shown on the display unit 23 of the user terminal 20. The lead agent 11b generates a prompt instructing the cost of the required cloud computing services and sends the prompt to the cost estimation agent 11cc. The cost estimation agent 11cc inputs the prompt to the artificial intelligence system 30, obtains the output result from the artificial intelligence system 30, and sends the output result to the lead agent 11b. As shown in Figure 17, the lead agent 11b displays a screen showing an estimated cost on the display unit 23 of the user terminal 20 based on the output result. The screen showing the estimated cost displays the total amount of the required cloud computing services (e.g., monthly), the cost of each cloud computing service, a breakdown by category, and cost optimization hints (suggestions for further cost reduction). The screen showing the estimated cost also displays the options to approve the proposal ("Approve and proceed") and to reject the proposal ("Reject and provide feedback"). If "Deny and provide feedback" is selected, the feedback input screen shown in Figure 10 will be displayed on the display unit 23 of the user terminal 20. 【0040】 Figure 18 shows an example of the configuration diagram generation input screen displayed on the display unit 23 of the user terminal 20. When "Create infrastructure configuration file and configuration diagram" is selected on the selection screen shown in Figure 6, the lead agent 11b displays the configuration diagram generation input screen on the display unit 23 of the user terminal 20, as shown in Figure 18. The configuration diagram generation input screen displays input fields for requirements information (system requirements) and a selection field for uploading an infrastructure configuration file. When the user enters system requirements in the requirements information (system requirements) input field, the artificial intelligence control unit 11a generates an infrastructure configuration file (see Figure 12) and a configuration diagram (see Figure 14) as proposed information for the cloud architecture and displays them on the display unit 23 of the user terminal 20. When an infrastructure configuration file is uploaded to the selection field for uploading an infrastructure configuration file, the artificial intelligence control unit 11a generates a configuration diagram (see Figure 14) as proposed information for the cloud architecture and displays it on the display unit 23 of the user terminal 20. 【0041】 Figure 19 shows an example of a cost estimate input screen displayed on the display unit 23 of the user terminal 20. When "Estimate cost" is selected on the selection screen shown in Figure 6, the lead agent 11b displays the cost estimate input screen on the display unit 23 of the user terminal 20, as shown in Figure 19. The cost estimate input screen displays input fields for requirements information (system requirements), a selection field for uploading an infrastructure configuration file, and a selection field for uploading a configuration diagram. When the user enters system requirements in the requirements information (system requirements) input field, or an infrastructure configuration file is uploaded to the selection field for uploading an infrastructure configuration file, or a configuration diagram is uploaded to the selection field for uploading a configuration diagram, the artificial intelligence control unit 11a generates the cost estimate result (see Figure 17) as proposed cloud architecture information and displays it on the display unit 23 of the user terminal 20. 【0042】 Figure 20 shows an example of the evaluation generation input screen displayed on the display unit 23 of the user terminal 20. When "Evaluate" is selected on the selection screen shown in Figure 6, the lead agent 11b displays the evaluation generation input screen on the display unit 23 of the user terminal 20, as shown in Figure 20. The evaluation generation input screen displays a selection field for uploading an infrastructure configuration file. When an infrastructure configuration file is uploaded to the selection field for uploading an infrastructure configuration file, the artificial intelligence control unit 11a generates the evaluation result (see Figure 16) as proposed information for the cloud architecture and displays it on the display unit 23 of the user terminal 20. 【0043】 [Differentiation] The embodiments described above are merely illustrative examples for implementing this disclosure. Therefore, this disclosure is not limited to the embodiments described above, and it is possible to implement the embodiments described above by modifying them as appropriate without departing from the spirit of the disclosure. 【0044】 (1) In the above embodiment, an example was shown in which the user enters text into the input field, but this disclosure is not limited thereto. For example, the design support system may be equipped with a control unit that functions as a chatbot, and the chatbot and the user may interact (the user enters text or voice input) to receive input from the user, or a tool (application program) provided by an external chat service provider may be used to allow the lead agent to obtain requirement information from the user through interaction with the user. 【0045】 (2) In the above embodiment, an example was shown in which an artificial intelligence system outside the design support system is used to acquire information from an external information management system, but this disclosure is not limited thereto. That is, an artificial intelligence system and an information management system may be built within the design support system. 【0046】 (3) In the above embodiments, an example of providing multiple artificial intelligence agents has been shown, but the disclosure is not limited thereto. That is, only a single artificial intelligence agent may be provided in the artificial intelligence control unit. 【0047】 (4) In the above embodiment, an example was shown in which the requirements input screen is provided with input fields for an overview and objectives, a budget, and performance requirements, but the disclosure is not limited thereto. For example, only at least one of the input fields for an overview and objectives, a budget, and performance requirements may be provided. 【0048】 (5) In the embodiments described above, the prompt has been shown to include an instruction on the perspective for analyzing the requirements information, an instruction on specifying the processing steps, an instruction on collecting information from multiple cloud computing services, and an instruction on outputting the results of the analysis of the requirements information, but the disclosure is not limited thereto. That is, the prompt does not have to include at least one of these instructions. 【0049】 (6) In the embodiments described above, the prompts were shown to include instructions for the artificial intelligence system to evaluate from the perspectives of operational excellence, security, reliability, performance efficiency, and cost optimization, but the disclosure is not limited thereto. That is, evaluations may be made from other perspectives. 【0050】 Furthermore, the above-described configuration can be explained as follows. 【0051】 The program relating to the first configuration is a program that causes the processor of the design support system to execute control processing. The program causes the processor to execute the following processes: acquiring requirement information in natural language input by the user from the user terminal; generating a prompt based on the requirement information that includes an instruction to output the necessary cloud computing services to the artificial intelligence system; inputting the prompt to the artificial intelligence system and acquiring the output result from the artificial intelligence system; generating proposed cloud architecture information based on the output result that proposes a cloud architecture which is design information of a combination of multiple cloud computing services; and sending the proposed cloud architecture information to the user terminal (first configuration). 【0052】 According to the first configuration described above, when a user inputs requirements information in natural language into their user terminal, proposed cloud architecture information is sent to the user terminal. This eliminates the need for the user to consider (design) combinations of multiple cloud computing services from a large number of cloud computing services, allowing the user to quickly obtain proposed cloud architectures. 【0053】 In the first configuration, the program may be configured to cause the processor to further execute a process to display at least one of the input fields for overview and purpose, budget, and performance requirements on the user terminal. The process for acquiring the requirements information may include a process for acquiring the information entered in at least one of the input fields for overview and purpose, budget, and performance requirements from the user terminal as the requirements information (second configuration). 【0054】 According to the second configuration described above, requirement information can be easily generated by the user entering information into the input fields displayed on the user terminal. 【0055】 In the first or second configuration, the process for generating the prompt may include instructions for analyzing the requirements information and instructions for perspectives on analyzing the requirements information, which include, in addition to the required cloud computing service, at least one of the following: an architectural pattern which is the structure of the required cloud computing service, security requirements, and cost optimization (third configuration). 【0056】 According to the third configuration described above, the artificial intelligence system can perform analysis on at least one of the following: architectural patterns, security requirements, and cost optimization, in addition to determining the necessary cloud computing services. Therefore, it is possible to provide users with suggested information based on at least one of these analyses. 【0057】 In any one of the first to third configurations, the process for generating the prompt may include a process for generating a prompt that includes an instruction to analyze the requirements information, an instruction to collect information on multiple cloud computing services, and an instruction to propose the necessary cloud computing services (fourth configuration). 【0058】 According to the fourth configuration described above, even if the artificial intelligence system has not learned information on multiple cloud computing services, it is possible to have the artificial intelligence system collect information on multiple cloud computing services and then propose the necessary cloud computing services. 【0059】 In any one of the first to fourth configurations, the process for generating the prompt may include a process for generating a prompt that includes an instruction to analyze the requirements information, an instruction to output the results of the analysis of the requirements information, and an instruction to propose the necessary cloud computing services (fifth configuration). 【0060】 According to the fifth configuration described above, the analysis results of the requirements information are output, allowing the user to check whether the analysis results of the requirements information match the requirements they have in mind. 【0061】 In any one of the first to fifth configurations, the process for generating the prompt may include a process for generating a prompt that includes an instruction to analyze the requirements information, an instruction to propose the necessary cloud computing services, and an instruction to output the cost of the necessary cloud computing services (sixth configuration). 【0062】 According to the sixth configuration described above, users can check the costs of the necessary cloud computing services. 【0063】 In any one of the first to sixth configurations, the program may be configured to further cause the processor to execute a process that generates an evaluation instruction prompt, which includes an instruction to cause the artificial intelligence system to evaluate the proposed information of the cloud architecture from the perspectives of operational excellence, security, reliability, performance efficiency, and cost optimization (seventh configuration). 【0064】 According to the seventh configuration described above, users can verify the operational excellence, security, reliability, performance efficiency, and cost optimization of the proposed grand architecture. 【0065】 In any one of the first to seventh configurations, the proposed cloud architecture information may include an image of the configuration diagram of the proposed cloud architecture. The process of sending the proposed cloud architecture information to the user terminal may include a process of displaying the image of the configuration diagram on the user terminal (eighth configuration). 【0066】 According to the eighth configuration described above, the user can view the configuration diagram of the proposed cloud architecture. 【0067】 In any one of the first to eighth configurations, the proposed cloud architecture information may include infrastructure configuration files for configuring the proposed cloud architecture. The process of sending the proposed cloud architecture information to the user terminal may include the process of sending the infrastructure configuration files to the user terminal (ninth configuration). 【0068】 According to the ninth configuration described above, the burden on users in creating infrastructure configuration files can be reduced. 【0069】 In any one of the first to ninth configurations, the process of sending the proposed cloud architecture information to the user terminal may include a process of displaying on the user terminal an option to approve the proposed cloud architecture and an option to reject the proposed cloud architecture, and if the rejection option is selected by the user, a process of displaying a feedback input field on the user terminal and acquiring the feedback information entered in the feedback input field. The program may further be configured to cause the processor to execute a process of generating a correction instruction prompt based on the feedback information, which includes an instruction to cause the artificial intelligence system to output the necessary cloud computing services again; a process of inputting the correction instruction prompt to the artificial intelligence system and acquiring the correction output result from the artificial intelligence system; a process of generating proposed correction information for the cloud architecture based on the correction output result; and a process of sending the proposed correction information for the cloud architecture to the user terminal (tenth configuration). 【0070】 According to the above-described configuration 10, even if the initial proposal information is not satisfactory to the user, based on the feedback information, proposal information that is likely to satisfy the user can be sent to the user's terminal. 【0071】 The design support system according to the 11th configuration includes: a requirements information acquisition unit that acquires requirements information in natural language input by the user from a user terminal; a prompt generation unit that generates a prompt including an instruction to the artificial intelligence system to output the necessary cloud computing services based on the requirements information; an output result acquisition unit that inputs the prompt to the artificial intelligence system and acquires the output result from the artificial intelligence system; a proposal information generation unit that generates proposed cloud architecture information, which is design information of a combination of multiple cloud computing services, based on the output result; and a transmission unit that transmits the proposed cloud architecture information to the user terminal (10th configuration). 【0072】 According to the above 11th configuration, it is possible to provide a design support system that can quickly propose a cloud architecture, which is design information for a combination of multiple cloud computing services, to the user. [Explanation of symbols] 【0073】 10: Server device, 11: Control unit, 11a: Artificial intelligence control unit, 11aa: Requirements information acquisition unit, 11ab: Prompt generation unit, 11ac: Output result acquisition unit, 11ad: Proposal information generation unit, 11ae: Transmission control unit, 11b: Read agent, 11c: Sub-agent, 11ca: Architect design agent, 11cb: Infrastructure configuration file generation agent, 11cc: Cost estimation agent, 11cd: Evaluation agent, 11d: Memory control unit, 12: Memory unit, 12a: Program, 12b: Deliverable database, 12c: History record database, 13: Communication unit, 20: User terminal, 21: Control unit, 22: Operation unit, 23: Display unit, 24: Communication unit, 25: Microphone, 26: Speaker, 27: Memory unit, 30: Artificial intelligence system, 100: Design support system, N: Network

Claims

[Claim 1] A program that causes the processor of a design support system to execute control processing so as to function as a plurality of artificial intelligence agents that perform the input of instructions to an artificial intelligence system and the acquisition of output from the artificial intelligence system on behalf of the user, The aforementioned multiple artificial intelligence agents include a lead agent and sub-agents, The aforementioned lead agent, The process involves obtaining requirements information in natural language entered by the user from the user terminal, The processor is instructed to perform a process that generates a prompt, based on the requirements information, which includes an instruction to output the necessary cloud computing services to the artificial intelligence system. The sub-agent causes the processor to execute a process that inputs the prompt to the artificial intelligence system and obtains the output result from the artificial intelligence system. The aforementioned lead agent, A process to generate proposed cloud architecture information, which is design information for a combination of multiple cloud computing services, based on the output results, A program that causes the processor to perform the process of sending the proposed cloud architecture information to the user terminal. [Claim 2] The lead agent further causes the processor to perform a process that displays at least one of the following on the user terminal: an input field for the overview and objective, an input field for the budget, and an input field for the performance requirements. The program according to claim 1, wherein the process for obtaining the requirements information includes a process for obtaining from the user terminal the information entered in at least one of the input fields for the overview and objectives, the input field for the budget, and the input field for the performance requirements, as the requirements information. [Claim 3] The process for generating the aforementioned prompt is: Instructions to analyze the aforementioned requirements information, The program according to claim 1, further comprising a process for generating a prompt, which includes, in addition to the necessary cloud computing service, instructions for a perspective when analyzing the requirements information, which include at least one of the following: an architectural pattern which is the structure of the necessary cloud computing service, security requirements, and cost optimization. [Claim 4] The process for generating the aforementioned prompt is: Instructions to analyze the aforementioned requirements information, Instructions to collect information from multiple cloud computing services, The program according to claim 1, comprising a process for generating a prompt, which includes an instruction to propose the necessary cloud computing service. [Claim 5] The process for generating the aforementioned prompt is: Instructions to analyze the aforementioned requirements information, An instruction to output the analysis results of the aforementioned requirements information, The program according to claim 1, comprising a process for generating a prompt, which includes an instruction to propose the necessary cloud computing service. [Claim 6] The process for generating the aforementioned prompt is: Instructions to analyze the aforementioned requirements information, Instructions to propose the necessary cloud computing services, The program according to claim 1, comprising a process for generating a prompt, which includes an instruction to output the cost of the necessary cloud computing service. [Claim 7] The program according to claim 1, wherein the lead agent further causes the processor to execute a process that generates an evaluation instruction prompt, which includes an instruction to cause the artificial intelligence system to evaluate the proposed information of the cloud architecture from the perspectives of operational excellence, security, reliability, performance efficiency, and cost optimization. [Claim 8] The proposed cloud architecture information includes an image of the configuration diagram of the proposed cloud architecture, The program according to claim 1, wherein the process of transmitting the proposed cloud architecture information to the user terminal includes the process of displaying the image of the configuration diagram on the user terminal. [Claim 9] A program for causing the processor of a design support system to execute control processing, The process involves obtaining requirements information in natural language entered by the user from the user terminal, A process that generates a prompt, including an instruction to output the necessary cloud computing services to the artificial intelligence system, based on the requirements information, The process of inputting the prompt into the artificial intelligence system and obtaining the output result from the artificial intelligence system, A process to generate proposed cloud architecture information, which is design information for a combination of multiple cloud computing services, based on the output results, The processor is instructed to perform the following: send the proposed cloud architecture information to the user terminal. The proposed cloud architecture information includes infrastructure configuration files for configuring the proposed cloud architecture, The process of sending the proposed cloud architecture information to the user terminal is a program that includes the process of sending the infrastructure configuration file to the user terminal. [Claim 10] A program for causing the processor of a design support system to execute control processing, The process involves obtaining requirements information in natural language entered by the user from the user terminal, A process that generates a prompt, including an instruction to output the necessary cloud computing services to the artificial intelligence system, based on the requirements information, The process of inputting the prompt into the artificial intelligence system and obtaining the output result from the artificial intelligence system, A process to generate proposed cloud architecture information, which is design information for a combination of multiple cloud computing services, based on the output results, The processor is instructed to perform the following: send the proposed cloud architecture information to the user terminal. The process of sending the proposed cloud architecture information to the user terminal is as follows: The user terminal is shown an option to approve the proposed cloud architecture and an option to reject the proposed cloud architecture. If the user selects the option to deny, a feedback input field will be displayed on the user's terminal. This process includes obtaining the feedback information entered in the feedback input field. The aforementioned program, A process that generates a correction instruction prompt, which includes an instruction to cause the artificial intelligence system to output the necessary cloud computing services again, based on the feedback information, The process involves inputting the correction instruction prompt to the artificial intelligence system and obtaining the correction output result from the artificial intelligence system. A process for generating proposed cloud architecture modifications based on the aforementioned modification output results, A program that causes the processor to perform a process of sending suggested modification information for the cloud architecture to the user terminal. [Claim 11] A plurality of artificial intelligence agents that perform on behalf of a user the input of instructions to an artificial intelligence system and the acquisition of output from the artificial intelligence system, the plurality of artificial intelligence agents including a lead agent and sub-agents, The aforementioned lead agent, A requirements information acquisition unit that acquires requirements information in natural language entered by the user from the user terminal, and It functions as a prompt generation unit that generates prompts, including instructions to output the necessary cloud computing services to the artificial intelligence system, based on the requirements information. The sub-agent functions as an output result acquisition unit that inputs the prompt to the artificial intelligence system and acquires the output result from the artificial intelligence system. The aforementioned lead agent, A proposal information generation unit generates cloud architecture proposal information, which is design information for a combination of multiple cloud computing services, based on the output results, and A design support system that further functions as a transmission unit for transmitting proposed information about the aforementioned cloud architecture to the user terminal.

Images