JP2013114437A - System construction support method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce work for system construction design by extracting a system configuration similar to that of a current project, out of system configurations of projects executed in the past to divert the extracted system configuration, in order to solve the problem that system configuration design work for determining the number of computers and software, their combinations, design of arrangement, computer collaboration systems, etc., is manually performed and takes time when an environment including developers, testers, and the like is constructed in a project of enterprise information system development.SOLUTION: A system construction support method includes: a project evaluation part 103 which calculates similarities between system requirements of a current project and those of past projects and man-hours of manual work executed by diverting computer environments of the past projects; and an input/output part 101 which extracts a past project candidate similar to the current project to display documented procedures for computer environment construction.

Description

  The present invention relates to the construction of a computing system. In particular, the present invention relates to a technology for constructing a computing system that executes a predetermined business including the financial field by using a result of examining past construction or construction.

  In the development of corporate information systems, it is necessary to construct various environments for each process, such as development machines and test machines, in addition to the production environment. Therefore, the subject that the environment construction takes a lot of time has occurred.

  As conventional techniques for solving this, there are Patent Literature 1 and Non-Patent Literature 1. In Patent Literature 1, environment construction time is reduced by automating the distribution and execution of an OS installation script. In Non-Patent Document 1, the environment construction time is reduced by creating a template combining an OS, a database, a TP monitor, and the like, and copying the disk image to another computer.

  With these technologies, if an operator creates an environment construction procedure manual for each computer, the environment can be automatically constructed according to the procedure manual, so that the environment construction time can be reduced.

Special table 2010-503074 gazette

IBM Tivoli Provisioning Manager http://www-06.ibm.com/software/jp/tivoli/products/prov_mng/

  However, in the prior art documents, the system configuration design of each process is still performed manually (manual work) and takes time. For example, there is a problem that it takes time and effort to design the number of computers and software, combination, arrangement, and design of a linkage system between computers.

  In order to solve the above problems, in the present invention, past projects similar to the project for the system to be constructed are identified, the identified projects are compared, and the result is constructed from the manual man-hours of the project. The manual man-hours in the system are calculated. Further, according to the present invention, it is possible to output a procedure manual indicating the procedure of construction based on the calculated manual man-hours, and to construct based on this, so that construction work can be executed more easily. Note that “similarity” refers to those having a relationship of a predetermined similarity (hereinafter). More specifically, the present invention has the following configuration.

  Comparing the current project and past project with a means to extract and display past project candidates similar to the current project, with a means to construct the environment of each computer based on the environment construction procedure manual created by the worker A means for calculating the similarity of system requirements, a means for comparing the current project with the past project and calculating an estimate of the manual labor when diverting the computer environment of the past project, and a computer environment construction procedure manual for the past project Is a system construction support method characterized by means for displaying.

  According to the present invention, it is possible to reduce the work of system configuration design by calculating a work amount or outputting a procedure manual.

It is a figure which shows the structure of the computer in 1st embodiment of this invention. It is a figure which shows the project list. It is a figure which shows the project group list. It is a figure which shows the environment construction instruction list 113. It is a figure which shows the system requirement similarity weighting table. It is a figure which shows the manual man-hour weighting table. It is a figure which shows the project evaluation result. It is a figure which shows the project group evaluation result 117. FIG. It is a flowchart explaining the process which receives the information of the present project, evaluates a similar past project, and displays a result. It is a figure which shows the information of the present project which the input / output part 101 receives from an operator. It is a flowchart explaining the process in which the project group evaluation part 102 calculates the system requirement similarity of a project group. It is a flowchart explaining the process in which the project group evaluation part 102 calculates the manual man-hour of a project group. It is a flowchart explaining the process in which the project evaluation part 103 calculates the system requirement similarity of a project. It is a flowchart explaining the process which the project evaluation part 103 calculates the manual man-hour of a project. It is a figure which shows the screen on which the input / output part 101 displays a result to an operator. It is a flowchart explaining the process which builds the environment of a computer according to the input environment construction instruction. It is a flowchart explaining the process which registers a past project.

  FIG. 1 is an overall view of the first embodiment. Computer 1 that displays information on past projects similar to the current project (the project that is going to build a system in the future), computer 2 that builds the environment of the computer according to the environment construction instruction, and the system development of the current project The computer 3 is connected through a network or the like. This configuration is an example of the first embodiment, and there is a function for displaying information on past projects similar to the current project in one computer and a function for building a computer environment according to an environment construction instruction. When it exists, there may be a case where a plurality of computers 1, 2, and 3 exist.

  A computer 1 that displays information on past projects similar to the current project is a computer including a CPU 10, a main memory 11, a storage device 12, an input device 13, an output device 14, and a communication interface 15. Connected by bus or the like. The CPU 10 performs various processes by reading a section (program) stored in the storage device 12 into the main memory 11 and executing it. In other words, each unit (101 to 104) can be realized as a so-called computer program (hereinafter the same applies to other computers). The storage device 12 includes an input / output unit 101 that manages data input from an operator and data output from a computer, and compares the information of the current project input from the operator with a group of past projects, and the similarity of system requirements. A project group evaluation unit 102 that calculates an approximation with manual labor, a project evaluation unit 103 that compares information of the current project with a past project, and calculates details of similarity of system requirements and manual labor, and a past project on the computer 1 Project information registration unit 104 for registering information on the past, a project list 111 for storing information on past projects, a project grouping list 112 for storing results of grouping past projects, and an environment for storing environment construction instructions used in past projects Construction instruction list 1 3. A system requirement similarity weighting table 114 storing weighting factors used when calculating system requirement similarity, a manual manhour weighting table 115 storing weighting factors used when calculating manual manhours, A project evaluation result 116 that stores a result of comparing and evaluating a past project and a current project, and a project group evaluation result 117 that stores a result of comparing and evaluating the project group and the current project are stored. Hereinafter, the project is also referred to as PJ.

  The computer 2 that constructs the environment of the computer in accordance with the environment construction instruction is a computer comprising a CPU 20, a main memory 21, a storage device 22, an input device 23, an output device 24, and a communication interface 25. Connected. The storage device 22 stores an environment construction instruction receiving unit 201 that receives an environment construction instruction input from an operator, an environment construction control unit 202 that transfers an appropriate disk image based on the received environment construction document, and a disk image. A disk image list 211 is stored.

  The computer 3 used in the system development of the current project is a computer comprising a CPU 30, a main memory 31, a storage device 32, an input device 33, an output device 34, and a communication interface 35, which are connected by an internal bus or the like. The The storage device 32 stores an environment construction execution unit 301 that constructs an environment using the disk image transferred from the environment construction control unit 202.

  In this embodiment, the computer environment is constructed by transferring and copying the disk image, but other computer environment construction means may be used.

  FIG. 2 is an example of the project list 111, and stores information on projects executed in the past. The project ID 1111 is an ID assigned so as to be unique for a project executed in the past. The system construction pattern 1112 is a standard pattern of the system infrastructure, and is a pattern defined in advance by an operator. In this embodiment, an OLTP (on-line transaction processing) system, a Web system, a batch system, a backup system, and the like are assumed. However, patterns may be divided from business viewpoints such as account management business and loan business. The system requirement 1113 is a requirement determined in each project. In this embodiment, continuity, fault tolerance, business throughput, and performance target values are taken as examples, but there are actually many items such as maintainability and safety. Levels 1 to 3 are set for each item of the system requirement, and the higher the value, the higher the requirement of the system requirement. For example, level 3 in continuity is to resume the business in less than 60 seconds when a failure occurs, and level 1 is to resume the business within 24 hours. The level of each system requirement is defined in advance by the worker. The software stack 1114 is the name and version information of the OS and middleware used in each project. In this embodiment, an OS and a database are taken as an example, but there are actually many software stacks such as TP monitor products and development environment products. The tool 1115 is the name of the tool used for each project and its version information. In this embodiment, the message input tool and the coverage tool are taken as examples, but there are actually many tools such as a file transfer tool and a data conversion tool. Each record corresponds to a project executed in the past, and one project is registered in FIG.

  FIG. 3 shows an example of the project grouping list 112, which stores the results of grouping past projects. The project group ID 1121 is an ID uniquely assigned to each project group. The system construction pattern 1122 is a system construction pattern of projects belonging to the project group. The system large requirement 1123 is a collection of project system requirements. In this embodiment, availability and performance are taken as an example, availability is an item that aggregates continuity and fault tolerance, and performance is an item that aggregates business processing volume and performance target value. Levels 1 to 3 are also set for each item of the system large requirement 1123. The higher the value, the higher the requirement of the system requirement. A software stack 1124 is a software stack of a project belonging to the project group. A tool 1125 is a tool used in a project belonging to a project group. Project ID 1126 is an ID of a project belonging to a project group. Each record corresponds to a grouped project group. In FIG. 3, one project group is registered.

  FIG. 4 shows an example of the environment construction instruction list 113, which stores environment construction instructions used in past projects. The project ID 1131 is an ID of a project that uses the instruction. The environment construction instruction 1132 is an environment construction instruction used in a past project, and is input information of the computer 2 that automatically constructs a computer environment. In this embodiment, it is described which disk image stored in the disk image list 211 is used to construct the environment.

  FIG. 5 is an example of the system requirement similarity weighting table 114, and stores weighting factors used when calculating the system requirement similarity. The system requirement 1141 stores items of the system requirement for weighting, and is an item of the system requirement 1113 and the system large requirement 1123. In the case of this embodiment, there are six items of continuity, fault tolerance, business processing volume, performance target value, availability, and performance. FIG. 5 describes a record relating to continuity among the six items. A system requirement similarity 1144 that is a weighting coefficient is defined by a combination of the current PJ requirement 1142 and the past PJ requirement 1143. For example, in the case of continuity, when the system requirement of the current project is level 3 and the system requirement of the past project is level 1, the weighting factor is 10. When the combination of the current PJ requirement 1142 and the past PJ requirement 1143 is not registered, the weight coefficient is set to 0.

  FIG. 6 shows an example of the manual man-hour weighting table 115, which stores a weighting factor used when calculating the manual man-hour. The work target 1151 stores items of software and tools to be weighted, and items of the software stack 1114 and the tool 1115 of the project list 111. In the case of this embodiment, there are four types: OS, DB, message input tool, and coverage tool. In FIG. 6, records related to the DB are described among the four items. Based on the combination of the current PJ name 1152, the current PJ Ver 1153, the past PJ name 1154, and the past PJ Ver 1155, a manual estimated man-hour 1157 that is a weighting factor and an update procedure 1156 that is the work content are extracted. For example, if the DB of the current project is DB_A and the version is 7.0, the DB of the past project is DB_A and the version is 8.0, the estimated manual man-hour is 1, and the update procedure is a manual update. Also, for each software stack and tool, a weighting factor is registered when a combination of the current PJ and past PJ versions is not registered or when the names of the current PJ and past PJ are different. For example, if the combination of version is not registered in the fifth record with the same DB, and the DB is different in the sixth record, the update procedure and manual estimated man-hours are registered.

  FIG. 7 shows an example of the project evaluation result 116, which stores the result of comparing and evaluating the past project and the current project. Project ID 1161 is an ID of a past project to be compared. The system requirement similarity (detail) 1162 is a detailed value of the similarity between the system requirements of the past project and the current project. The manual labor (details) 1163 is a detailed value of the manual labor required when diverting a past project. These values are calculated by the project evaluation unit 103. In FIG. 7, the evaluation results of five projects are stored.

  FIG. 8 shows an example of the project group evaluation result 117, which stores the result of comparing and evaluating the project group and the current project. A project group ID 1171 is an ID of a project group to be compared. The system requirement similarity (approximate) 1172 is an approximate value of the similarity between the system requirements of the project group and the current project. The manual man-hour (estimated) 1173 is an approximate value of the manual man-hours required when diverting past projects belonging to the project group. These values are calculated by the project evaluation unit 103. These values are calculated by the project group evaluation unit 102. In FIG. 8, the evaluation results of three project groups are stored.

Next, the flow of processing in this embodiment will be described.
First, a process in which an operator inputs information on the current project to the computer 1, evaluates a past project in which the computer 1 is registered, and displays the evaluation result to the operator will be described with reference to the flowchart of FIG. To do.

  First, in S11, the input from the operator is received via the input / output unit 101, and information on the current project is registered. An example of information to be registered is shown in FIG. In FIG. 10, the system construction pattern to be constructed in the current project, the system requirements expected in the current project, the name and version of the software stack to be installed, and the name and version of the tool to be used in the development process and test process are shown. Stored.

  In S12, the project group evaluation unit 102 compares the input current project information with the project group list 112, and calculates the system requirement similarity (approximate) of each project group. A detailed flowchart of S12 is shown in FIG.

  First, the system requirements for the input current project are calculated (S121). The value of the system large requirement is the maximum value among the aggregated system requirements. For example, if the continuity level is 3 and the fault tolerance level is 2, the availability level is 3. Next, one project group whose system construction pattern 1112 matches the current project is extracted from the project group list 112 (S122). Next, the system large requirement items of the project group and the current project are compared (S123). If the compared items are different, the system requirement similarity weighting table 114 is referred to, and the value of the corresponding system requirement similarity 1144 is acquired (S124). After comparing the large system requirements (S125), the project group evaluation result 117 is updated with the project group ID 1171 and the system requirement similarity (estimated) 1172 (S126). The value of the system requirement similarity (approximate) is a value obtained by adding up the values of the system requirement similarity acquired in S124. After comparing each project group (S127), the process proceeds to the next step.

  In S <b> 13, the project group evaluation unit 102 compares the input current project information with the project group list 112 and calculates the manual man-hours (approximate) of each project group. A detailed flowchart of S13 is shown in FIG.

  First, one project group whose system construction pattern 1112 matches the current project is extracted from the project group list 112 (S131). Next, the software stack and tool items of the project group and the current project are compared (S132). If the compared items are different, the manual estimated man-hour estimation table 115 is referred to, and the value of the corresponding manual estimated man-hour 1157 is acquired (S133). After comparing all the software stacks and tools (S134), the project group ID 1171 and the manual man-hour (estimated) 1173 are updated to the project group evaluation result 117 (S135). The value of the manual man-hour (approximate) is the sum of the values of the manual labor estimated man-hour acquired in S133. After comparing all project groups (S136), the process proceeds to the next step.

  In S <b> 14, the project group evaluation unit 102 refers to the project group evaluation result 117, and selects project candidates for performing detailed evaluation. A candidate having a high system requirement similarity (approximate) 1172 and manual man-hour (approximate 1173) is selected as a candidate. As an example of the selection method, either the system requirement similarity (approximate) 1172 or manual man-hour (approximate 1173) corresponds to the top 100% of all projects, or either value corresponds to the top 100 Is mentioned.

  In S15, the project evaluation unit 103 compares each project selected in S14 with the current project, and calculates the system requirement similarity (details). A detailed flowchart of S15 is shown in FIG.

  One project to be compared is extracted (S151), and the system requirement items of the project and the current project are compared (S152). If the compared items are different, the system requirement similarity weighting table 114 is referred to, and the value of the corresponding system requirement similarity 1144 is acquired (S153). After comparing the system requirements (S154), the project evaluation result 116 is updated with the project ID 1161 and the system requirement similarity (details) 1162 (S155). The value of the system requirement similarity (detail) 1162 is the sum of the values of the system requirement similarity obtained in S153. The projects are compared (S156), and the process proceeds to the next step (S16).

  In S16, the project evaluation unit 103 compares each project selected in S14 with the current project, and calculates the manual man-hours (details). A detailed flowchart of S16 is shown in FIG.

  One project to be compared is extracted (S161), and the software stack of the project and the current project are compared with the name and version of the tool (S162). If the compared name and version are different, the manual man-hour estimation table 115 is referred to, and the value of the corresponding manual man-hour estimation man-hour 1157 is acquired (S163). After comparing all the software and tools (S164), the project ID 1161 and the manual man-hour (detail) 1163 are updated to the project evaluation result 116 (S165). The value of manual labor (details) 1163 is a value obtained by adding up the values of the manual labor estimated labor obtained in S163. The projects are compared (S166), and the process proceeds to the next step (S17).

  In S17, the input / output unit 101 refers to the project evaluation result 116 and the environment construction instruction list 113, and displays the result to the operator. An example of the display is shown in FIG.

  FIG. 15 shows the top three projects with the same system requirement similarity and the top three projects with the manual labor, and displays the system requirement similarity with each project, the manual labor, and the environment construction instructions used in the project. doing. The worker confirms the displayed project and determines which project environment construction instruction is to be diverted.

  Second, a process in which the operator inputs the result displayed on the computer 1 to the computer 2 and the computer 2 constructs the environment of the computer 3 will be described with reference to the flowchart of FIG. Here, although the top three cases are used, this may be other cases such as two cases, three or more cases, or the number of cases having a predetermined similarity or more.

  In S21, the environment construction instruction receiving unit 201 receives the environment construction procedure manual 1132 from the worker. For example, when it is determined that the worker constructs the system by diverting the P001 project, the worker registers four environment construction procedure manuals SE001, SE002, SE003, and SE004 in the environment construction instruction reception part. In S 22, the environment construction control unit 202 refers to the corresponding disk image from the disk image list 211 according to the environment construction procedure manual 1132 and sends it to the environment construction execution unit 301. In S23, the disk image sent by the environment construction execution unit 301 is copied to the storage device 32.

  Thirdly, a process for registering information of a project executed by a worker in the past in the computer 1 will be described with reference to a flowchart of FIG.

  First, the project information registration unit 104 receives information on a project executed in the past from the worker (S31). The information input by the worker is the project information shown in FIG. 10 and the environment construction procedure manual used in the project. Specifically, the system construction pattern, system requirements, software stack name and version, tool name and version, and environment construction procedure used in the project.

  Next, the project information registration unit 104 adds the received project ID, and then adds a record to the project list 111 (S32). Next, the environment construction procedure manual used in the project and the corresponding project ID are registered in the environment construction instruction list 113 (S33). Next, the system requirements for the input project information are calculated (S34). The value of the system large requirement is the maximum value among the aggregated system requirements. For example, if the continuity level is 3 and the fault tolerance level is 2, the availability level is 3. Next, for all the project groups stored in the project group list 112, the system large requirement, the name of the software stack, and the name of the tool are compared, and it is confirmed whether there is a project group that matches all ( S35). If it does not exist, a new project group is created and added to the project group list 112 (S36). If it exists, the project ID of the project ID 1126 in the project group list 112 is updated (S37).

101 I / O unit 102 Project group evaluation unit 103 Project evaluation unit 104 Project information registration unit 111 Project list 112 Project group list 113 Environment construction instruction list 114 System requirement similarity weighting table 115 Manual man-hour weighting table 116 Project evaluation result 117 Project Group Evaluation Result 201 Environment Construction Instruction Accepting Unit 202 Environment Construction Control Unit 211 Disk Image List 301 Environment Construction Execution Unit

Claims (8)

A system construction support apparatus that supports the system construction in the development of a system that executes a predetermined task,
A method for building the environment of each computer based on the environment construction procedure manual created by the worker;
Means for comparing current and past projects for the system;
Means for calculating the similarity of the system requirements of the system using the result of the comparison;
And a means for calculating an estimate of manual man-hours when the computer environment of the past project is diverted using the result of the comparison. The system construction support device according to claim 1,
A system construction support apparatus further comprising means for outputting a computer environment construction procedure manual for the past project based on the estimate. In the system construction support device according to claim 1 or 2,
The system construction support apparatus characterized in that the means for calculating the estimate estimates the manual man-hour from a software stack and a tool name and version installed in the system of the past project. The system construction support device according to any one of claims 1 to 3,
Further comprising means for grouping the past projects based on system requirements of the past projects and installed software stacks and tools;
The comparing means compares the grouped information with the information of the current project. A system construction support method for supporting the system construction in the development of a system that executes a predetermined job,
Building the environment of each computer based on the environment construction procedure manual created by the worker;
Comparing a current project and a past project for the system;
Calculating the similarity of the system requirements of the system using the result of the comparison;
And calculating an estimate of manual man-hours when the computer environment of the past project is diverted using the result of the comparison. The system construction support method according to claim 5,
A system construction support method further comprising a step of outputting a computer environment construction procedure manual for the past project based on the estimate. In the system construction support method according to claim 5 or 6,
The step of calculating the estimate comprises estimating the manual man-hour from a software stack and a tool name and version installed in the system of the past project. The system construction support method according to any one of claims 5 to 7,
Further comprising grouping the past projects based on the system requirements of the past projects and installed software stacks and tools;
The comparing step includes comparing the grouped information with the information of the current project.

Images