WO2014068979A1 - Performance estimating system, perform ance estimating method, and program - Google Patents

Performance estimating system, perform ance estimating method, and program Download PDF

Info

Publication number
WO2014068979A1
WO2014068979A1 PCT/JP2013/006436 JP2013006436W WO2014068979A1 WO 2014068979 A1 WO2014068979 A1 WO 2014068979A1 JP 2013006436 W JP2013006436 W JP 2013006436W WO 2014068979 A1 WO2014068979 A1 WO 2014068979A1
Authority
WO
WIPO (PCT)
Prior art keywords
model
performance
function
processing
application
Prior art date
Application number
PCT/JP2013/006436
Other languages
French (fr)
Japanese (ja)
Inventor
一男 矢野尾
隆夫 大崎
Original Assignee
日本電気株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 日本電気株式会社 filed Critical 日本電気株式会社
Priority to JP2014544321A priority Critical patent/JPWO2014068979A1/en
Publication of WO2014068979A1 publication Critical patent/WO2014068979A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
    • G06F11/3447—Performance evaluation by modeling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/30—Monitoring
    • G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
    • G06F11/3409—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment for performance assessment
    • G06F11/3414—Workload generation, e.g. scripts, playback

Definitions

  • the present invention relates to a performance estimation device, a performance estimation method, and a program related to an IT (Information Technology) system.
  • the configuration of the IT system and the scale of the IT system are estimated even in a situation where the detailed functions of the application are not determined, and the price of the IT system is estimated. May be determined. Even in such a case, if an IT system similar to the IT system has already been operated, whether the IT system satisfies the required performance based on information related to the performance of the similar IT system. Estimate whether or not. Evaluating whether the IT system meets the required performance reduces the chances of a project that designs the IT system failing.
  • Non-Patent Document 1 estimates the performance of an IT system.
  • Non-Patent Document 2 discloses a technique for applying a model-based technique to design related to an IT system. The technology verifies whether the designed IT system satisfies the required performance (for example, function and availability) from the initial stage of design.
  • Patent Document 1 discloses a performance evaluation apparatus using such MARTE.
  • Patent Document 1 in order to estimate the performance related to the IT system, the user directly describes the parameters in the system model representing the platform of the IT system.
  • upstream SE system engineer designing an IT system in an upstream process to find a place where necessary parameters are described in the system model.
  • the upstream SE hears the requirements of the system from the customer and proposes an IT system based on the found requirements.
  • the upstream SE estimates a parameter related to the execution time and the like based on information close to the function of the application, such as the number of input / output files and the number of accesses to the database, and past examples.
  • the upstream SE may not fully know such knowledge and experience. Therefore, it is difficult to evaluate the performance related to the model base in the upstream process. Furthermore, it is difficult to divert system models to different IT systems.
  • ⁇ By describing information related to processing performed by the application in the system model, more accurate performance can be estimated. For example, when an application includes a process for acquiring an object (for example, lock, critical section, semaphore, etc.) necessary for synchronization and a process for releasing the acquired object, or includes parallel processing, an object to be synchronized ( Hereinafter, it is expressed as “synchronous object”) and parallel processing is described in the system model.
  • an object to be synchronized hereinafter, it is expressed as “synchronous object” and parallel processing is described in the system model.
  • a main object of the present invention is to provide a performance estimation device and the like that facilitate system design.
  • a performance estimation device is characterized by comprising the following configuration.
  • the performance estimation device is: A system model including information related to a platform included in the IT system, and an AP model including information related to AP-specific processing that is processing specific to an application that realizes the function in the IT system and information related to parameters included in the application are common.
  • Model synthesis means for creating a performance model by synthesizing the system model and the AP model based on the message
  • estimating means for calculating the performance of the IT system according to a predetermined calculation procedure based on the performance model.
  • the performance estimation method is: In the information processing apparatus, a system model including information related to a platform included in the IT system, an AP including information related to an AP specific process that is a process specific to an application that realizes a function in the IT system, and information related to a parameter included in the application Based on the message that the model includes in common, the system model and the AP model are combined to create a performance model, and based on the performance model, the performance in the IT system is determined according to a predetermined calculation procedure. It is characterized by calculating.
  • this object is also realized by such a performance estimation program and a computer-readable recording medium for recording the program.
  • an IT system performance estimation device that can be easily used for an upstream SE and that models can be easily reused. That is, according to the performance estimation device and the like according to the present invention, it is possible to provide a performance estimation device and the like that facilitate system design.
  • FIG. 1 is a block diagram showing a configuration of a performance estimation apparatus 10 according to an embodiment of the present invention.
  • the performance estimation device 10 includes a model selection unit 101, an AP parameter input unit 102, a model synthesis unit 103, an estimation unit 104, a model storage unit 105, and an output device 106.
  • the performance estimation device 10 is a central processing unit (Central Processing Unit, hereinafter referred to as “CPU”), a read-only memory (Read Only Memory, hereinafter referred to as “ROM”), or a random access memory (Random Access Memory).
  • CPU Central Processing Unit
  • ROM Read Only Memory
  • RAM Random Access Memory
  • RAM Random Access Memory
  • the performance estimation device 10 may be configured by a single computer, but may have a plurality of computers connected to each other via a communication line (communication network, hereinafter simply referred to as “network”). There may be.
  • network communication network
  • the model selection unit 101, the AP parameter input unit 102, the model synthesis unit 103, and the estimation unit 104 correspond to modules representing functions realized by the CPU executing a predetermined program stored in a ROM or the like.
  • the model storage unit 105 can be realized by an external storage device (not shown). The external storage device may be connected to the performance estimation device 10 via the performance estimation device 10 and a network.
  • the output device 106 is, for example, a display device such as a display or a printer, and outputs various images and characters in response to a signal output from the CPU of the performance estimation device 10.
  • FIG. 10 illustrates a configuration example of hardware resources that implements the performance estimation device according to each embodiment of the present invention described above using one calculation processing device (information processing device, computer).
  • FIG. 10 is a diagram schematically showing a hardware configuration of a calculation processing apparatus capable of realizing a performance estimation apparatus according to each embodiment of the present invention to be described later.
  • the calculation processing device 20 includes a CPU 21, a memory 22, a disk 23, a nonvolatile recording medium 24, an input device 25, and an output device 26.
  • the nonvolatile recording medium 24 can be read by a computer, for example, a compact disc (Compact Disc), a digital versatile disc (Digital_Versatile_Disc), a Blu-ray Disc (Blu-ray Disc. Registered trademark), a universal serial bus memory (USB memory), This refers to a solid state drive, etc., which retains the program without power supply and enables it to be carried.
  • the nonvolatile recording medium 24 is not limited to the above-described medium. Further, the program may be carried via a communication network instead of the nonvolatile recording medium 24.
  • the CPU 21 copies a software program (computer program: hereinafter simply referred to as “program”) stored in the disk 23 to the memory 22 and executes arithmetic processing.
  • the CPU 21 reads data necessary for program execution from the memory 22. When the display is necessary, the CPU 21 displays the output result on the output device 26. When inputting a program from the outside, the CPU 21 reads the program from the input device 25.
  • the CPU 21 interprets and executes the performance estimation device control program (FIG. 6) in the memory 22 corresponding to the function (process) represented by each unit shown in FIG.
  • the CPU 21 sequentially performs the processes described in the above-described embodiments of the present invention.
  • the present invention can also be realized by such a performance estimation device control program. Furthermore, it can be understood that the present invention can also be realized by a computer-readable non-volatile recording medium in which the performance estimation device control program is recorded.
  • the model storage unit 105 can store a plurality of system models and a plurality of application models (hereinafter referred to as “AP models”).
  • the system model includes information related to the components included in the platform in the IT system and information related to operations between the components.
  • the AP model includes a parameter in the application (hereinafter referred to as “AP parameter”) and information related to processing unique to the application.
  • the model selection unit 101 selects a necessary system model and AP model from the model storage unit 105.
  • the AP parameter input unit 102 acquires an AP parameter included in the selected AP model.
  • the model synthesizing unit 103 synthesizes a model for estimating the performance related to the IT system to be evaluated, using the system model and AP model selected by the model selecting unit 101 and the AP parameter acquired by the AP parameter input unit 102.
  • the estimation unit 104 estimates the performance related to the IT system based on the model synthesized by the model synthesis unit 103.
  • the output device 106 outputs the result estimated by the estimation unit 104.
  • FIGS. 2A and 2B are diagrams illustrating examples of system models.
  • the system model includes a server diagram, a process diagram, and a function flow diagram.
  • the server diagram is a diagram showing a configuration related to a platform (hereinafter also referred to as “hardware”) included in the IT system.
  • the server diagram includes two servers “WebServer” and “DBServer” and one client PC, a communication network switch (hereinafter referred to as “network switch”), and a router. It represents connecting.
  • the process diagram represents the connection relationship between processes that realize the application.
  • the process diagram represents a relationship in which three processes “Apache”, “Tomcat”, and “MySQL” are connected in series.
  • the server diagram and process diagram can be described using a composite structure diagram (Composite Structure Diagram) in the Unified Modeling Language (hereinafter referred to as “UML”) as shown in FIGS. 2A and 2B. it can.
  • “WebServer: Server” in the rectangle represents “Server class role WebServer”.
  • the server diagram and the process diagram may be described using an internal block diagram (Internal Block Diagram) in a system modeling language (Systems Modeling Language, hereinafter referred to as “SysML”).
  • SynsML Systems Modeling Language
  • the server diagram and the process diagram may be described using an object diagram in UML.
  • a dotted arrow in FIG. 2B represents which element (ie, hardware) in the server diagram executes the element (ie, process) in the process diagram. This can be described using “allocation” defined by SysML or MARTE.
  • the class related to the elements in the process diagram represents the “function” that the process has.
  • the function flow diagram represents a sequence (hereinafter referred to as “execution sequence”) that realizes the “function” of the process.
  • execution sequence a sequence that realizes the “function” of the process.
  • the functional flow diagram defines two execution sequences.
  • the functional flow can be identified, for example, using the name that the initial message has.
  • the first functional flow defined by the functional flow diagram is “Static browsing”.
  • “Static browsing” represents a sequence for making a request for browsing static content from the client function to the World Wide Web (World Wide Web, hereinafter referred to as “WWW”) function and receiving the requested result.
  • WWW World Wide Web
  • the second functional flow defined by the functional flow diagram is “dynamic processing”. “Dynamic processing” represents an execution sequence shown in dynamic processing 1 to dynamic processing 3, for example.
  • the WWW function requests the AP function to perform processing in response to a request for browsing dynamic content from the client function to the WWW function.
  • the AP function inquires about a database to a database (database, hereinafter referred to as “DB”) function in response to a request.
  • DB database
  • the AP function returns the content related to the dynamically calculated content to the WWW function based on the response performed by the DB.
  • a functional flow diagram can be described using a collaboration diagram (Collaboration diagram or Communication diagram) in UML.
  • the functional flow diagram is described using a collaboration diagram, but the functional flow diagram may be described using a sequence diagram or an activity diagram.
  • branch processing, parallel processing, and the like can be described.
  • the relationship between the elements in the process diagram and the elements in the function flow diagram is defined based on the class related to the element in the process diagram. You may describe using allocation like the relationship between the elements in a figure.
  • FIG. 3 is a diagram illustrating an example of an AP model.
  • the AP model associates AP parameters, parameter conversion expressions, AP functions, and the like for each workload.
  • the workload represents a series of processes (or also expressed as “functional flow”, “sequence”, etc.) at the application level.
  • Each workload is associated with a functional flow that can realize the workload.
  • the AP model includes “purchase processing”, “dynamic processing”, “data size, AP time, and background processing time”, and function A (a sequence diagram related to function A in FIG. 4). To represent.)
  • AP parameters are design parameters at the application level.
  • the creator of the AP model that is, the application developer
  • the parameter conversion formula is a calculation formula that converts the content related to the AP parameter into the parameter in the system model.
  • the creator of the AP model defines a parameter conversion formula according to the target application.
  • the workload “HP display” includes the AP parameter “data size”.
  • the AP parameter “data size” represents an average of data sizes of HTML (HyperText Markup Language) content to be displayed.
  • the parameter conversion formula represents, for example, the service time per transaction in the “static browsing” process as a primary formula of the AP parameter “data size”. In the example shown in FIG. 3, the parameter conversion formula represents that the coefficient in the linear formula is 0.01.
  • the service time related to the “static browsing” process is the CPU usage time per transaction when the reference server processes the static browsing.
  • the workload “product browsing” includes the AP parameter “AP processing load” in addition to the AP parameter “data size”.
  • the AP parameter “AP processing load” is a parameter for estimating the CPU processing time of “AP processing” per transaction. In the parameter conversion formula, a high value, a medium value, or a low value can be set for the AP parameter “AP processing load”.
  • the AP parameter “AP processing load” “high” indicates, for example, that the CPU processing time is 100 milliseconds (hereinafter referred to as “ms”).
  • ms milliseconds
  • the CPU processing time is 50 ms.
  • the AP parameter “AP processing load” is “medium”, for example, the CPU processing time is 5 ms.
  • the user determines in advance whether the “AP processing load” is high, medium, or low. This method is easier than when the CPU processing time is explicitly estimated.
  • the parameter conversion formula in the workload “purchase processing” indicates that the service time per transaction in the AP function is equal to the AP parameter “AP processing time”.
  • the upstream SE inputs the AP parameter “AP processing time” by explicitly estimating the CPU processing time per transaction of the AP function.
  • AP specific processing represents processing performed by a function in a certain workload.
  • FIG. 4 is a sequence diagram showing AP-specific processing in the “AP function” when the workload is “purchase processing”.
  • the AP server generates a work thread while synchronously executing an inquiry to a database (hereinafter referred to as “DB”), and in parallel with the generated work thread, Work in parallel (hereinafter simply referred to as “parallel”).
  • DB database
  • the messages “AP processing”, “DB inquiry”, “result”, and “DB result” in FIG. 4 are the messages “AP processing” in the functional flow diagrams shown in FIGS. 2A and 2B, respectively. It is associated with “DB inquiry”, “result”, and “DB result”.
  • the AP-specific process is described using a sequence diagram, but may be described using a collaboration diagram or an activity diagram.
  • FIGS. 5A to 5C are diagrams illustrating an example of a use case in the embodiment of the present invention.
  • A Platform architect who creates a system model and registers the created system model in the repository
  • B An application developer who creates an AP model and registers the created AP model in the repository
  • C An upstream SE that searches a model by referring to a repository, estimates the performance of an application based on the detected model, and uses the estimated performance for system proposal to a customer.
  • the platform architect creates a system model with the same functional flow diagram for the system model that can be considered the same at the functional level.
  • the platform architect creates two different system models “Web 3-tier system model (OSS)” and “Web 3-tier system model (high availability)” having the same functional flow diagram. To do.
  • an application developer can create an AP model without considering the details of the system model other than the function flow diagram. Furthermore, the platform architect can accumulate system models without considering the application.
  • the upstream SE selects an AP model and a system model, and sets AP parameters related to each workload included in the selected AP model. Thereby, the upstream SE can estimate the performance of the IT system without knowing the details of the application and the platform.
  • the upstream SE selects an application to be proposed to the customer via the model selection unit 101. Thereafter, the model selection unit 101 displays a list of system models that can realize the application. Specifically, the model selection unit 101 displays a list of system models by searching for system models that include all processing flows associated with the AP model.
  • the upstream SE inputs AP parameters according to the circumstances specific to the case via the AP parameter input unit 102.
  • the model combining unit 103 refers to a model (hereinafter referred to as “performance model”) that is referred to when estimating the performance of the IT system based on the AP parameter input by the upstream SE, the selected system model, and the AP model. Is generated.
  • the estimation unit 104 displays, on the output device 106, the result of estimating the performance of the parameters input by the upstream SE based on the performance model generated by the model synthesis unit 103 according to a predetermined calculation procedure.
  • the estimation unit 104 may calculate an estimated value by analytically calculating a performance model, or may calculate an estimated value by simulation.
  • the estimation unit 104 may be realized using a technique disclosed in Japanese Patent Application No. 2011-132377 (hereinafter referred to as “Patent Document 2”).
  • the model composition unit 103 will be described with reference to an example of generating a performance model related to the workload “purchase processing”. To do.
  • FIG. 7 is a diagram illustrating a state in which an execution flow at the function level is generated when the converted execution flow is represented by an activity diagram.
  • the process flow is described using a collaboration diagram, and the AP-specific process is described using a sequence diagram. Both are associated via a message name.
  • the process of generating the execution flow at the function level can be realized according to generation procedure 1 to generation procedure 3.
  • Generation Procedure 1 An execution flow (hereinafter referred to as “first execution flow”) is generated by associating a message with a node (that is, a node in the graph theory) in the order of messages in the processing flow.
  • Generation Procedure 2 An execution flow (hereinafter referred to as “second execution flow”) is generated by associating a message with a node for the processing content in the AP-specific processing.
  • the content related to the sequence diagram may be interpreted using, for example, the method described in Reference 2, and may be represented using an activity diagram equivalent to the content related to the interpreted sequence diagram.
  • Generation Procedure 3 The first execution flow and the second execution flow are synthesized.
  • the node in the first execution flow and the node in the second execution flow are associated with the message name. Therefore, in the synthesis process, the nodes associated with the same message name are regarded as the same node in the first execution flow and the second execution flow.
  • parameters at the function level are calculated based on the AP parameters input to the AP parameter input unit 102 and the parameter conversion formula associated with the AP parameters (step S2).
  • the input AP parameter is expressed as follows.
  • step S3 the execution flow at the function level is converted into an execution flow at the process level.
  • functions and processes are associated one to one. For this reason, this conversion processing can be realized by replacing a node in the execution flow at the function level with a node representing a process associated with the function. Note that even when functions and processes are associated in a many-to-one relationship, processing can be performed in a similar manner when associated in a one-to-one relationship.
  • step S4 the execution flow at the process level is converted into a performance model.
  • An example of this processing is described in Patent Document 2, for example. For this reason, in this embodiment, the detailed description regarding the process to convert is abbreviate
  • the upstream SE can input parameters significantly easily. This is because the information about the platform is hidden in the upstream SE, and the upstream SE only needs to set the AP parameter.
  • the AP model and the system model are determined first.
  • the AP model may be determined and the AP parameter may be input first, and then the system model may be selected.
  • the proposed IT system does not agree with the user in terms of price and performance as a result of estimation, re-estimation can be performed by changing only the system model without changing the AP model and AP parameter input results. It becomes easy to do.
  • the function flow diagram included in the system model and the AP specific process included in the AP model are combined.
  • the processing sequence functional flow
  • the processing sequence may be described in the AP model without describing the processing sequence in the system model.
  • FIG. 8 For example, a system model is described as shown in FIG. 8 and an AP model is described as shown in FIG. 9A, which is an example of a system model that does not include these and has a function diagram that defines functions. Describe the relationship between functions and processes using allocation.
  • FIG. 9A instead of AP-specific processing, an execution sequence at a function level in the workload is described. In such a configuration, the AP model creator can easily model the AP processing in more detail.
  • the client function requests static browsing to the WWW function.
  • the WWW function performs processing in response to the request, and transmits the processing result to the client function.
  • the performance estimation apparatus facilitates system design.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Biology (AREA)
  • Quality & Reliability (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • Debugging And Monitoring (AREA)
  • Stored Programmes (AREA)

Abstract

The present invention discloses a performance estimating system, etc. for facilitating syst em design. The performance estimating system has a model synthesizing unit and an estimating unit. In the model synthesizing unit, a system model containing information on a platform of an IT system, and an AP model containing info rmation on an AP-specific process, which is a process specific to an application for realizing a function in the IT system and information on a parameter included in the application, are combined on the basis of a message included in the system model and the AP model alike to create a performance model. In the e stimating unit, performance in the IT system is calculated on the basis of the performance model, according to a prescribed calculation procedure.

Description

性能見積もり装置、性能見積もり方法、及びプログラムPerformance estimation device, performance estimation method, and program
　本発明は、ＩＴ（Ｉｎｆｏｒｍａｔｉｏｎ　Ｔｅｃｈｎｏｌｏｇｙ）システムに関する性能見積もり装置、性能見積もり方法、及びプログラムに関する。 The present invention relates to a performance estimation device, a performance estimation method, and a program related to an IT (Information Technology) system.
　ＵＭＬ（Ｕｎｉｆｉｅｄ　Ｍｏｄｅｌｉｎｇ　Ｌａｎｇｕａｇｅ）等のモデリング言語を用いて記述されたシステム・ソフトウェアに関するモデルに基づいて、ＩＴシステムに関する性能（たとえば、最大スループット、平均応答時間等）を見積もる技術（すなわち、モデルに基づく性能評価技術）が知られている。この技術によれば、ＩＴシステムを開発する初期段階から該ＩＴシステムに関する性能を見積もることによって、該ＩＴシステムが有する機能を実現するアプリケーションソフトウェア（以降、「アプリケーション」と記す。）が必要な性能を満たすか否かを検証できる。 Technology for estimating IT system performance (for example, maximum throughput, average response time, etc.) based on a system software model described using a modeling language such as UML (Unified Modeling Language) (ie, model-based performance) Evaluation technology) is known. According to this technology, the performance required for application software (hereinafter referred to as “application”) that realizes the functions of the IT system by estimating the performance of the IT system from the initial stage of developing the IT system. It can be verified whether it satisfies.
　また、顧客に新しいＩＴシステムを提案する際には、アプリケーションが有する詳細な機能を決定していない状況においても、該ＩＴシステムが有する構成及び該ＩＴシステムが有する規模を見積もり、該ＩＴシステムの価格を決定する場合がある。このような場合にも、該ＩＴシステムに類似するＩＴシステムが既に運用されている場合には、類似するＩＴシステムが有する性能に関する情報などに基づいて、該ＩＴシステムが要求された性能を満たすか否かを見積もる。該ＩＴシステムが要求された性能を満たすか否かを評価することにより、該ＩＴシステムを設計するプロジェクトが失敗する可能性を低減する。 In addition, when proposing a new IT system to a customer, the configuration of the IT system and the scale of the IT system are estimated even in a situation where the detailed functions of the application are not determined, and the price of the IT system is estimated. May be determined. Even in such a case, if an IT system similar to the IT system has already been operated, whether the IT system satisfies the required performance based on information related to the performance of the similar IT system. Estimate whether or not. Evaluating whether the IT system meets the required performance reduces the chances of a project that designs the IT system failing.
　たとえば、非特許文献１が開示するＭＡＲＴＥ（Ｍｏｄｅｌｉｎｇ　ａｎｄ　Ａｎａｌｙｓｉｓ　ｏｆ　Ｒｅａｌ－Ｔｉｍｅ　Ｅｍｂｅｄｄｅｄ　Ｓｙｓｔｅｍｓ）は、ＩＴシステムに関する性能を見積もる。また、非特許文献２は、モデルに基づく技術を、ＩＴシステムに関する設計に適用する技術を開示する。該技術は、設計の初期段階から、設計するＩＴシステムが要求された性能（たとえば、機能、及び、可用性等）を満たすか否かを検証する。 For example, MARTE (Modeling and Analysis of Real-Time Embedded Systems) disclosed in Non-Patent Document 1 estimates the performance of an IT system. Non-Patent Document 2 discloses a technique for applying a model-based technique to design related to an IT system. The technology verifies whether the designed IT system satisfies the required performance (for example, function and availability) from the initial stage of design.
　また、特許文献１は、係るＭＡＲＴＥを用いる性能評価装置を開示する。 Further, Patent Document 1 discloses a performance evaluation apparatus using such MARTE.
特開２００７－１８８１７９号公報JP 2007-188179 A
　しかし、特許文献１並びに非特許文献１及び２が開示する技術においては、ＩＴシステムに関する性能を見積もるために、ユーザは、パラメータを、ＩＴシステムが有するプラットフォームを表すシステムモデルに、直接記述する。 However, in the techniques disclosed in Patent Document 1 and Non-Patent Documents 1 and 2, in order to estimate the performance related to the IT system, the user directly describes the parameters in the system model representing the platform of the IT system.
　このため、上流工程においてＩＴシステムを設計するシステムエンジニア（上流ＳＥ）は、該システムモデルにおいて、必要なパラメータが記述されている場所を見つけることが難しいこともある。上流ＳＥは、顧客からシステムが有する要件を聞き出し、聞き出した要件に基づいて、ＩＴシステムを提案する。 For this reason, it may be difficult for a system engineer (upstream SE) designing an IT system in an upstream process to find a place where necessary parameters are described in the system model. The upstream SE hears the requirements of the system from the customer and proposes an IT system based on the found requirements.
　さらに、ＩＴシステムを提案する段階や、該ＩＴシステムを開発する初期の段階においては、該ＩＴシステムが有する機能を実現するアプリケーションは存在しない。このため、上流ＳＥは、アプリケーションを用いて、実行時間等に関するパラメータの値を得ることはできない。この場合、上流ＳＥは、入出力するファイル数や、データベースにアクセスする回数等、アプリケーションが有する機能に近い情報と、過去における事例とに基づいて、実行時間等に関するパラメータを推測する。 Furthermore, at the stage of proposing an IT system and the initial stage of developing the IT system, there are no applications that realize the functions of the IT system. For this reason, the upstream SE cannot obtain the value of the parameter relating to the execution time or the like using the application. In this case, the upstream SE estimates a parameter related to the execution time and the like based on information close to the function of the application, such as the number of input / output files and the number of accesses to the database, and past examples.
　パラメータを推測するためには、アプリケーションおよびプラットフォームに関する詳細な知識と経験が必要である。しかし、上流ＳＥは、そのような知識や経験について十分に知らないこともある。したがって、上流工程においてモデルベースに関する性能を評価することは難しい。さらに、システムモデルを異なるＩＴシステムに転用することは難しい。 推測 Detailed knowledge and experience about the application and platform are required to infer the parameters. However, the upstream SE may not fully know such knowledge and experience. Therefore, it is difficult to evaluate the performance related to the model base in the upstream process. Furthermore, it is difficult to divert system models to different IT systems.
　アプリケーションが行う処理に関する情報を、システムモデルに記述することにより、さらに正確な性能を見積もることができる。たとえば、アプリケーションが同期に必要なオブジェクト（たとえば、ロック、クリティカルセクション、セマフォ等）を取得する処理、及び、取得したオブジェクトを解放する処理を含む場合や、並列処理を含む場合に、同期するオブジェクト（以降、「同期オブジェクト」と表す。）や並列処理をシステムモデルに記述する。システムモデルに記述することにより、同期オブジェクトがボトルネックになるか否か、または、並列処理において同期することが原因となり応答時間が増大するか否か等について考察することができる。 ∙ By describing information related to processing performed by the application in the system model, more accurate performance can be estimated. For example, when an application includes a process for acquiring an object (for example, lock, critical section, semaphore, etc.) necessary for synchronization and a process for releasing the acquired object, or includes parallel processing, an object to be synchronized ( Hereinafter, it is expressed as “synchronous object”) and parallel processing is described in the system model. By describing in the system model, it is possible to consider whether the synchronization object becomes a bottleneck or whether the response time increases due to synchronization in parallel processing.
　しかし、個々のアプリケーションに特有な情報をシステムモデルに記述する場合には、該アプリケーションに関するシステムモデルを、異なるアプリケーションに転用することが難しい。 However, when information specific to each application is described in the system model, it is difficult to divert the system model related to the application to a different application.
　そこで、本発明は、上流ＳＥにも利用しやすく、モデルの再利用もしやすいＩＴシステムの性能見積もり装置を提供することを一つの目的とする。すなわち、本発明の主たる目的は、システム設計を容易にする性能見積もり装置等を提供することである。 Therefore, it is an object of the present invention to provide an IT system performance estimation device that can be easily used for upstream SEs and that models can be easily reused. That is, a main object of the present invention is to provide a performance estimation device and the like that facilitate system design.
　前述の目的を達成するために、本発明に係る性能見積もり装置は、以下の構成を備えることを特徴とする。 In order to achieve the above object, a performance estimation device according to the present invention is characterized by comprising the following configuration.
　すなわち、本発明に係る性能見積もり装置は、
　ＩＴシステムが有するプラットフォームに関する情報を含むシステムモデルと、前記ＩＴシステムにおける機能を実現するアプリケーションに固有な処理であるＡＰ固有処理に関する情報及び前記アプリケーションに含まれるパラメータに関する情報を含むＡＰモデルとが、共通して含むメッセージに基づき、前記システムモデルと前記ＡＰモデルとを合成することにより、性能モデルを作成するモデル合成手段と、
　前記性能モデルに基づき、所定の算出手順に従い、前記ＩＴシステムにおける性能を算出する見積り手段と
　を備えることを特徴とする。 That is, the performance estimation device according to the present invention is:
A system model including information related to a platform included in the IT system, and an AP model including information related to AP-specific processing that is processing specific to an application that realizes the function in the IT system and information related to parameters included in the application are common. Model synthesis means for creating a performance model by synthesizing the system model and the AP model based on the message
And estimating means for calculating the performance of the IT system according to a predetermined calculation procedure based on the performance model.
　また、本発明の他の見地として、本発明に係る性能見積もり方法は、
　情報処理装置において、ＩＴシステムが有するプラットフォームに関する情報を含むシステムモデルと、前記ＩＴシステムにおける機能を実現するアプリケーションに固有な処理であるＡＰ固有処理に関する情報及び前記アプリケーションに含まれるパラメータに関する情報を含むＡＰモデルとが、共通して含むメッセージに基づき、前記システムモデルと前記ＡＰモデルとを合成することにより、性能モデルを作成し、前記性能モデルに基づき、所定の算出手順に従い、前記ＩＴシステムにおける性能を算出する
　ことを特徴とする。 As another aspect of the present invention, the performance estimation method according to the present invention is:
In the information processing apparatus, a system model including information related to a platform included in the IT system, an AP including information related to an AP specific process that is a process specific to an application that realizes a function in the IT system, and information related to a parameter included in the application Based on the message that the model includes in common, the system model and the AP model are combined to create a performance model, and based on the performance model, the performance in the IT system is determined according to a predetermined calculation procedure. It is characterized by calculating.
　さらに、同目的は、係る性能見積もりプログラム、および、そのプログラムを記録するコンピュータ読み取り可能な記録媒体によっても実現される。 Furthermore, this object is also realized by such a performance estimation program and a computer-readable recording medium for recording the program.
　本発明によれば、上流ＳＥにも利用しやすく、モデルの再利用もしやすいＩＴシステムの性能見積もり装置を提供することができる。すなわち、本発明に係る性能見積もり装置等によれば、システム設計を容易にする性能見積もり装置等を提供することができる。 According to the present invention, it is possible to provide an IT system performance estimation device that can be easily used for an upstream SE and that models can be easily reused. That is, according to the performance estimation device and the like according to the present invention, it is possible to provide a performance estimation device and the like that facilitate system design.
本発明の実施の形態に係る性能見積もり装置が有する構成を示すブロック図である。It is a block diagram which shows the structure which the performance estimation apparatus which concerns on embodiment of this invention has. システムモデルの例を表す図である。It is a figure showing the example of a system model. システムモデルの例を表す図である。It is a figure showing the example of a system model. 本実施の形態による、ＡＰモデルの例を示す図である。It is a figure which shows the example of AP model by this Embodiment. 本実施の形態による、ＡＰ固有処理の例を示す図である。It is a figure which shows the example of AP specific process by this Embodiment. 本実施の形態におけるユースケースについて説明する図である。It is a figure explaining the use case in this Embodiment. 本実施の形態におけるユースケースについて説明する図である。It is a figure explaining the use case in this Embodiment. 本実施の形態におけるユースケースについて説明する図である。It is a figure explaining the use case in this Embodiment. 本実施の形態による、モデル合成部の動作に関するフローチャートである。It is a flowchart regarding operation | movement of the model synthetic | combination part by this Embodiment. 本実施の形態による、機能レベルの実行フロー生成を説明する図である。It is a figure explaining the execution flow production | generation of a function level by this Embodiment. 本実施の形態による、システムモデルの他の例を示す図である。It is a figure which shows the other example of the system model by this Embodiment. 本実施の形態による、ＡＰモデルの一例を示す図である。It is a figure which shows an example of AP model by this Embodiment. 本実施の形態による、ＡＰモデルの一例を示す図である。It is a figure which shows an example of AP model by this Embodiment. 本実施の形態による、ＡＰモデルの一例を示す図である。It is a figure which shows an example of AP model by this Embodiment. 本実施の形態による、ＡＰモデルの一例を示す図である。It is a figure which shows an example of AP model by this Embodiment. 本発明の各実施形態に係る性能見積もり装置を実現可能な計算処理装置のハードウェア構成を、概略的に示すブロック図である。It is a block diagram which shows roughly the hardware constitutions of the calculation processing apparatus which can implement | achieve the performance estimation apparatus which concerns on each embodiment of this invention.
　この出願は、２０１２年１１月２日に出願された日本出願特願２０１２－２４２６０６を基礎とする優先権を主張し、その開示の全てをここに取り込む。そして、本国際出願では、説明の便宜上から、係る日本出願の図面の表現を、国際出願の規定を満足すべく複数の図面に展開することとする。これに応じて、本出願人は、以降に説明する各実施形態の説明も、調整後の図面に応じた記載に調整することとする。これらの処置は、言語の違いに起因する、各国移行時の言語変換への配慮であり、何ら新規事項を含む処置ではない。 This application claims priority based on Japanese Patent Application No. 2012-242606 filed on Nov. 2, 2012, the entire disclosure of which is incorporated herein. In this international application, for the convenience of explanation, the representation of the drawings of the Japanese application will be developed into a plurality of drawings to satisfy the provisions of the international application. Accordingly, the applicant will adjust the description of each embodiment described below to the description according to the adjusted drawing. These measures are considerations for language conversion at the time of transition from one country to another due to language differences, and are not measures that include any new matter.
　以下、本発明を実施するための形態について、図面を参照して詳細に説明する。 Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the drawings.
　図１は、本発明の実施の形態に係る性能見積もり装置１０が有する構成を示すブロック図である。図１に示すように、性能見積もり装置１０は、モデル選択部１０１、ＡＰパラメータ入力部１０２、モデル合成部１０３、見積もり部１０４、モデル記憶部１０５、及び、出力装置１０６を有する。 FIG. 1 is a block diagram showing a configuration of a performance estimation apparatus 10 according to an embodiment of the present invention. As shown in FIG. 1, the performance estimation device 10 includes a model selection unit 101, an AP parameter input unit 102, a model synthesis unit 103, an estimation unit 104, a model storage unit 105, and an output device 106.
　性能見積もり装置１０は、中央処理演算装置（Ｃｅｎｔｒａｌ　Ｐｒｏｃｅｓｓｉｎｇ　Ｕｎｉｔ、以降「ＣＰＵ」と表す。）、リードオンリーメモリ（Ｒｅａｄ　Ｏｎｌｙ　Ｍｅｍｏｒｙ、以降「ＲＯＭ」と表す。）、または、ランダムアクセスメモリ（Ｒａｎｄｏｍ　Ａｃｃｅｓｓ　Ｍｅｍｏｒｙ、以降「ＲＡＭ」と表す。）等の記憶装置、外部記憶装置、入力インタフェース、出力インタフェース、通信インタフェース及び各装置を結ぶバス（すなわち、「ｂｕｓ」）を有する専用、または、汎用のコンピュータを用いて実現できる。 The performance estimation device 10 is a central processing unit (Central Processing Unit, hereinafter referred to as “CPU”), a read-only memory (Read Only Memory, hereinafter referred to as “ROM”), or a random access memory (Random Access Memory). Hereinafter referred to as “RAM”), etc., using a dedicated or general-purpose computer having a storage device such as a storage device, an external storage device, an input interface, an output interface, a communication interface, and a bus (ie, “bus”) connecting the devices. Can be realized.
　なお、性能見積もり装置１０は、単一のコンピュータにより構成されるものであっても、通信回線（通信ネットワーク、以降、単に「ネットワーク」と表す）を介して互いに接続する複数のコンピュータを有する構成であってもよい。 The performance estimation device 10 may be configured by a single computer, but may have a plurality of computers connected to each other via a communication line (communication network, hereinafter simply referred to as “network”). There may be.
　モデル選択部１０１、ＡＰパラメータ入力部１０２、モデル合成部１０３、見積もり部１０４は、ＣＰＵがＲＯＭ等に格納されている所定のプログラムを実行することにより実現する機能を表すモジュールに相当する。モデル記憶部１０５は、外部記憶装置（不図示）により実現できる。外部記憶装置は、性能見積もり装置１０及びネットワーク等を介して、性能見積もり装置１０と接続してもよい。 The model selection unit 101, the AP parameter input unit 102, the model synthesis unit 103, and the estimation unit 104 correspond to modules representing functions realized by the CPU executing a predetermined program stored in a ROM or the like. The model storage unit 105 can be realized by an external storage device (not shown). The external storage device may be connected to the performance estimation device 10 via the performance estimation device 10 and a network.
　出力装置１０６は、たとえば、ディスプレイ等の表示装置やプリンタであり、性能見積もり装置１０が有するＣＰＵが出力する信号を受けて、各種画像や文字を出力する。 The output device 106 is, for example, a display device such as a display or a printer, and outputs various images and characters in response to a signal output from the CPU of the performance estimation device 10.
　図１０に、上述した本発明の各実施形態における性能見積もり装置を、１つの計算処理装置（情報処理装置、コンピュータ）を用いて実現するハードウェア資源の構成例について説明する。図１０は、後述する本発明の各実施形態に係る性能見積もり装置を実現可能な計算処理装置のハードウェア構成を概略的に示す図である。 FIG. 10 illustrates a configuration example of hardware resources that implements the performance estimation device according to each embodiment of the present invention described above using one calculation processing device (information processing device, computer). FIG. 10 is a diagram schematically showing a hardware configuration of a calculation processing apparatus capable of realizing a performance estimation apparatus according to each embodiment of the present invention to be described later.
　計算処理装置２０は、ＣＰＵ２１、メモリ２２、ディスク２３、不揮発性記録媒体２４、入力装置２５、および、出力装置２６を有する。 The calculation processing device 20 includes a CPU 21, a memory 22, a disk 23, a nonvolatile recording medium 24, an input device 25, and an output device 26.
　不揮発性記録媒体２４は、コンピュータが読み取り可能な、たとえば、コンパクトディスク（Ｃｏｍｐａｃｔ　Ｄｉｓｃ）、デジタルバーサタイルディスク（Ｄｉｇｉｔａｌ＿Ｖｅｒｓａｔｉｌｅ＿Ｄｉｓｃ）、ブルーレイディスク（Ｂｌｕ－ｒａｙ　Ｄｉｓｃ。登録商標）、ユニバーサルシリアルバスメモリ（ＵＳＢメモリ）、ソリッドステートドライブ（Ｓｏｌｉｄ　Ｓｔａｔｅ　Ｄｒｉｖｅ）等を指しており、電源を供給しなくても係るプログラムを保持し、持ち運びを可能にする。不揮発性記録媒体２４は、上述した媒体に限定されない。また、不揮発性記録媒体２４の代わりに、通信ネットワークを介して係るプログラムを持ち運びしてもよい。 The nonvolatile recording medium 24 can be read by a computer, for example, a compact disc (Compact Disc), a digital versatile disc (Digital_Versatile_Disc), a Blu-ray Disc (Blu-ray Disc. Registered trademark), a universal serial bus memory (USB memory), This refers to a solid state drive, etc., which retains the program without power supply and enables it to be carried. The nonvolatile recording medium 24 is not limited to the above-described medium. Further, the program may be carried via a communication network instead of the nonvolatile recording medium 24.
　すなわち、ＣＰＵ２１は、ディスク２３が記憶するソフトウェア・プログラム（コンピュータ・プログラム：以下、単に「プログラム」と称する）を、実行する際にメモリ２２にコピーし、演算処理を実行する。ＣＰＵ２１は、プログラム実行に必要なデータをメモリ２２から読み取る。表示が必要な場合には、ＣＰＵ２１は、出力装置２６に出力結果を表示する。外部からプログラムを入力する場合、ＣＰＵ２１は、入力装置２５からプログラムを読み取る。ＣＰＵ２１は、上述した図１に示した各部が表す機能（処理）に対応するところのメモリ２２にある性能見積もり装置制御プログラム（図６）を解釈し実行する。ＣＰＵ２１は、上述した本発明の各実施形態において説明した処理を順次行う。 That is, the CPU 21 copies a software program (computer program: hereinafter simply referred to as “program”) stored in the disk 23 to the memory 22 and executes arithmetic processing. The CPU 21 reads data necessary for program execution from the memory 22. When the display is necessary, the CPU 21 displays the output result on the output device 26. When inputting a program from the outside, the CPU 21 reads the program from the input device 25. The CPU 21 interprets and executes the performance estimation device control program (FIG. 6) in the memory 22 corresponding to the function (process) represented by each unit shown in FIG. The CPU 21 sequentially performs the processes described in the above-described embodiments of the present invention.
　すなわち、このような場合、本発明は、係る性能見積もり装置制御プログラムによっても成し得ると捉えることができる。更に、係る性能見積もり装置制御プログラムが記録されたコンピュータ読み取り可能な不揮発性の記録媒体によっても、本発明は成し得ると捉えることができる。 That is, in such a case, it can be understood that the present invention can also be realized by such a performance estimation device control program. Furthermore, it can be understood that the present invention can also be realized by a computer-readable non-volatile recording medium in which the performance estimation device control program is recorded.
　モデル記憶部１０５は、複数のシステムモデルと、複数のアプリケーションモデル（以降、「ＡＰモデル」と表す。）とを記憶できる。 The model storage unit 105 can store a plurality of system models and a plurality of application models (hereinafter referred to as “AP models”).
　システムモデルは、ＩＴシステムにおけるプラットフォームが有する構成要素に関する情報と、各構成要素間における動作に関する情報とを含む。ＡＰモデルは、アプリケーションにおけるパラメータ（以降、「ＡＰパラメータ」と表す。）及び該アプリケーションに固有な処理に関する情報を含む。 The system model includes information related to the components included in the platform in the IT system and information related to operations between the components. The AP model includes a parameter in the application (hereinafter referred to as “AP parameter”) and information related to processing unique to the application.
　まず、モデル選択部１０１は、モデル記憶部１０５から必要なシステムモデルとＡＰモデルとを選択する。ＡＰパラメータ入力部１０２は、選択したＡＰモデルが含むＡＰパラメータを取得する。 First, the model selection unit 101 selects a necessary system model and AP model from the model storage unit 105. The AP parameter input unit 102 acquires an AP parameter included in the selected AP model.
　モデル合成部１０３は、モデル選択部１０１が選択するシステムモデル及びＡＰモデル、及び、ＡＰパラメータ入力部１０２が取得したＡＰパラメータを用いて、評価対象であるＩＴシステムに関する性能を見積もるモデルを合成する。見積もり部１０４は、モデル合成部１０３が合成するモデルに基づいて、ＩＴシステムに関する性能を見積もる。出力装置１０６は、見積もり部１０４が見積もった結果を出力する。 The model synthesizing unit 103 synthesizes a model for estimating the performance related to the IT system to be evaluated, using the system model and AP model selected by the model selecting unit 101 and the AP parameter acquired by the AP parameter input unit 102. The estimation unit 104 estimates the performance related to the IT system based on the model synthesized by the model synthesis unit 103. The output device 106 outputs the result estimated by the estimation unit 104.
　図２Ａ及び図２Ｂは、システムモデルの例を表す図である。図２Ａ及び図２Ｂを参照すると、システムモデルは、サーバ図、プロセス図、及び、機能フロー図を含む。サーバ図は、ＩＴシステムが有するプラットフォーム（以降、「ハードウェア」とも表現する。）に関する構成を表す図である。 2A and 2B are diagrams illustrating examples of system models. 2A and 2B, the system model includes a server diagram, a process diagram, and a function flow diagram. The server diagram is a diagram showing a configuration related to a platform (hereinafter also referred to as “hardware”) included in the IT system.
　図２Ｂに示す例において、サーバ図は、「ＷｅｂＳｅｒｖｅｒ」、「ＤＢＳｅｒｖｅｒ」という２台のサーバと１台のクライアントＰＣとを、通信ネットワークスイッチ（以降、「ネットワークスイッチ」と表す。）とルータとを介して接続することを表す。 In the example illustrated in FIG. 2B, the server diagram includes two servers “WebServer” and “DBServer” and one client PC, a communication network switch (hereinafter referred to as “network switch”), and a router. It represents connecting.
　プロセス図は、アプリケーションを実現するプロセス間の接続関係を表す。図２Ｂの例において、プロセス図は、「Ａｐａｃｈｅ」、「Ｔｏｍｃａｔ」、「ＭｙＳＱＬ」なる３つのプロセスが直列に接続する関係を表す。 The process diagram represents the connection relationship between processes that realize the application. In the example of FIG. 2B, the process diagram represents a relationship in which three processes “Apache”, “Tomcat”, and “MySQL” are connected in series.
　サーバ図とプロセス図は、図２Ａ及び図２Ｂに示すように、統一モデリング言語（Ｕｎｉｆｉｅｄ　Ｍｏｄｅｌｉｎｇ　Ｌａｎｇｕａｇｅ、以降、「ＵＭＬ」と表す。）におけるコンポジット構成図（Ｃｏｍｐｏｓｉｔｅ　Ｓｔｒｕｃｔｕｒｅ　Ｄｉａｇｒａｍ）を用いて記述することができる。ここで、矩形の中の「ＷｅｂＳｅｒｖｅｒ：Ｓｅｒｖｅｒ」は、「ＳｅｒｖｅｒクラスのロールＷｅｂＳｅｒｖｅｒ」を表す。なお、サーバ図とプロセス図とを、システムモデリング言語（Ｓｙｓｔｅｍｓ　Ｍｏｄｅｌｉｎｇ　Ｌａｎｇｕａｇｅ、以降、「ＳｙｓＭＬ」と表す）における内部ブロック図（Ｉｎｔｅｒｎａｌ　Ｂｌｏｃｋ　Ｄｉａｇｒａｍ）を用いて記述してもよい。あるいは、サーバ図とプロセス図とを、ＵＭＬにおけるオブジェクト図を用いて記述してもよい。 The server diagram and process diagram can be described using a composite structure diagram (Composite Structure Diagram) in the Unified Modeling Language (hereinafter referred to as “UML”) as shown in FIGS. 2A and 2B. it can. Here, “WebServer: Server” in the rectangle represents “Server class role WebServer”. The server diagram and the process diagram may be described using an internal block diagram (Internal Block Diagram) in a system modeling language (Systems Modeling Language, hereinafter referred to as “SysML”). Alternatively, the server diagram and the process diagram may be described using an object diagram in UML.
　図２Ｂにおける点線矢印は、サーバ図中のどの要素（すなわち、ハードウェア）が、プロセス図中の要素（すなわち、プロセス）を実行するかを表す。これは、ＳｙｓＭＬまたはＭＡＲＴＥが規定する「アロケーション」を用いて記述できる。 A dotted arrow in FIG. 2B represents which element (ie, hardware) in the server diagram executes the element (ie, process) in the process diagram. This can be described using “allocation” defined by SysML or MARTE.
　プロセス図中の要素に関するクラスは、プロセスが有する「機能」を表す。 The class related to the elements in the process diagram represents the “function” that the process has.
　ここで、機能フロー図は、プロセスが有する「機能」を実現するシーケンス（以降、「実行シーケンス」と表す、）を表す。図２Ａ及び図２Ｂが示す例において、機能フロー図は、２つの実行シーケンスを定義する。機能フローは、たとえば、最初のメッセージが有する名前を用いて識別することができる。 Here, the function flow diagram represents a sequence (hereinafter referred to as “execution sequence”) that realizes the “function” of the process. In the example shown in FIGS. 2A and 2B, the functional flow diagram defines two execution sequences. The functional flow can be identified, for example, using the name that the initial message has.
　機能フロー図が定義する１つ目の機能フローは「静的閲覧」である。「静的閲覧」は、クライアント機能からワールドワイドウェブ（Ｗｏｒｌｄ　Ｗｉｄｅ　Ｗｅｂ、以降、「ＷＷＷ」と表す。）機能に静的コンテンツを閲覧する要求を行い、要求した結果を受け取るシーケンスを表す。 The first functional flow defined by the functional flow diagram is “Static browsing”. “Static browsing” represents a sequence for making a request for browsing static content from the client function to the World Wide Web (World Wide Web, hereinafter referred to as “WWW”) function and receiving the requested result.
　機能フロー図が定義する２つ目の機能フローは「動的処理」である。「動的処理」は、たとえば、動的処理１乃至動的処理３に示す実行シーケンスを表す。 The second functional flow defined by the functional flow diagram is “dynamic processing”. “Dynamic processing” represents an execution sequence shown in dynamic processing 1 to dynamic processing 3, for example.
　　（動的処理１）ＷＷＷ機能は、クライアント機能からＷＷＷ機能に動的コンテンツを閲覧する要求に応じて、ＡＰ機能に処理を依頼する、
　　（動的処理２）ＡＰ機能は、依頼に応じて、データベース（ｄａｔａｂａｓｅ、以降「ＤＢ」と表す）機能にデータベースに関して問い合わせる、
　　（動的処理３）ＡＰ機能は、ＤＢが行う応答に基づいて、動的に算出するコンテンツに関する内容をＷＷＷ機能に返す。 (Dynamic processing 1) The WWW function requests the AP function to perform processing in response to a request for browsing dynamic content from the client function to the WWW function.
(Dynamic processing 2) The AP function inquires about a database to a database (database, hereinafter referred to as “DB”) function in response to a request.
(Dynamic processing 3) The AP function returns the content related to the dynamically calculated content to the WWW function based on the response performed by the DB.
　図２Ａ及び図２Ｂに示すように、ＵＭＬにおけるコラボレーション図（Ｃｏｌｌａｂｏｒａｔｉｏｎ　ｄｉａｇｒａｍ、あるいは、コミュニケーション図（Ｃｏｍｍｕｎｉｃａｔｉｏｎ　Ｄｉａｇｒａｍ）））を用いて、機能フロー図を記述することができる。本実施形態においては、機能フロー図を、コラボレーション図を用いて記述するが、機能フロー図を、シーケンス図またはアクティビティ図を用いて記述してもよい。たとえば、機能フロー図を、シーケンス図やアクティビティ図を用いて記述することにより、分岐処理や並列処理等も記述することができる。 As shown in FIGS. 2A and 2B, a functional flow diagram can be described using a collaboration diagram (Collaboration diagram or Communication diagram) in UML. In the present embodiment, the functional flow diagram is described using a collaboration diagram, but the functional flow diagram may be described using a sequence diagram or an activity diagram. For example, by describing a functional flow diagram using a sequence diagram or activity diagram, branch processing, parallel processing, and the like can be described.
　また、本実施形態においては、プロセス図中の要素に関するクラスに基づいて、プロセス図中の要素と、機能フロー図中の要素との間の関係を定義するが、プロセス図中の要素と、サーバ図中の要素との間の関係のように、アロケーションを用いて記述してもよい。 In this embodiment, the relationship between the elements in the process diagram and the elements in the function flow diagram is defined based on the class related to the element in the process diagram. You may describe using allocation like the relationship between the elements in a figure.
　図３は、ＡＰモデルの例を表す図である。ＡＰモデルは、ワークロードごとに、ＡＰパラメータ、パラメータ変換式、及び、ＡＰ機能等を関連付ける。本実施形態において、ワークロードは、アプリケーションレベルにおける一連の処理（または、「機能フロー」、「シーケンス」等とも表す。）を表す。各ワークロードは、そのワークロードを実現可能な機能フローに関連づけされている。 FIG. 3 is a diagram illustrating an example of an AP model. The AP model associates AP parameters, parameter conversion expressions, AP functions, and the like for each workload. In the present embodiment, the workload represents a series of processes (or also expressed as “functional flow”, “sequence”, etc.) at the application level. Each workload is associated with a functional flow that can realize the workload.
　たとえば、図３において、ＡＰモデルは、「購入処理」と、「動的処理」と、「データサイズ、ＡＰ時間、及び、背景処理時間」と、機能Ａ（図４に機能Ａに関するシーケンス図を表す。）とを関連付ける。 For example, in FIG. 3, the AP model includes “purchase processing”, “dynamic processing”, “data size, AP time, and background processing time”, and function A (a sequence diagram related to function A in FIG. 4). To represent.)
　ＡＰパラメータは、アプリケーションレベルにおける設計パラメータである。ＡＰモデルの作成者（すなわち、アプリケーション開発者）は、対象とするアプリケーションに関する個々の事情に応じてＡＰパラメータを定義する。 AP parameters are design parameters at the application level. The creator of the AP model (that is, the application developer) defines the AP parameters according to individual circumstances regarding the target application.
　パラメータ変換式は、ＡＰパラメータに関する内容を、システムモデル中のパラメータに変換する計算式である。ＡＰモデルの作成者は、対象となるアプリケーションに応じて、パラメータ変換式を定義する。 The parameter conversion formula is a calculation formula that converts the content related to the AP parameter into the parameter in the system model. The creator of the AP model defines a parameter conversion formula according to the target application.
　図３に示す例においては、ワークロード「ＨＰ表示」は、ＡＰパラメータ「データサイズ」を含む。ＡＰパラメータ「データサイズ」は、表示するＨＴＭＬ（ＨｙｐｅｒＴｅｘｔ　Ｍａｒｋｕｐ　Ｌａｎｇｕａｇｅ）コンテンツが有するデータサイズの平均を表す。また、パラメータ変換式は、たとえば、「静的閲覧」処理における１トランザクションあたりのサービス時間をＡＰパラメータ「データサイズ」の一次式で表す。図３に示す例において、パラメータ変換式は、該一次式における係数が０．０１であることを表す。 In the example shown in FIG. 3, the workload “HP display” includes the AP parameter “data size”. The AP parameter “data size” represents an average of data sizes of HTML (HyperText Markup Language) content to be displayed. The parameter conversion formula represents, for example, the service time per transaction in the “static browsing” process as a primary formula of the AP parameter “data size”. In the example shown in FIG. 3, the parameter conversion formula represents that the coefficient in the linear formula is 0.01.
　ここで、「静的閲覧」処理に関するサービス時間は、基準となるサーバが静的閲覧を処理する場合の１トランザクション当たりのＣＰＵ使用時間である。 Here, the service time related to the “static browsing” process is the CPU usage time per transaction when the reference server processes the static browsing.
　ＣＰＵ使用時間を正確に得るためには、実装したアプリケーションを用いて測定する必要がある。しかし、該アプリケーションを提案する段階においては、該アプリケーションの実装は完了していない。この場合には、過去の事例や測定値に基づき、エキスパートが目安となる計算式を作成してもよい。 In order to accurately obtain the CPU usage time, it is necessary to measure using the installed application. However, the implementation of the application is not completed at the stage of proposing the application. In this case, based on past cases and measured values, a calculation formula for an expert may be created.
　過去の事例等に基づいて作成された計算式は、提案時においてＩＴシステムが有する規模や価格等を見積もる場合には有用である。 [Formulas created based on past cases, etc. are useful when estimating the scale, price, etc. of IT systems at the time of proposal.
　ワークロード「商品閲覧」は、ＡＰパラメータ「データサイズ」の他にＡＰパラメータ「ＡＰ処理負荷」を含む。ＡＰパラメータ「ＡＰ処理負荷」は、１トランザクション当たりの「ＡＰ処理」のＣＰＵ処理時間を見積もる場合におけるパラメータである。パラメータ変換式は、ＡＰパラメータ「ＡＰ処理負荷」に高、中、または、低のいずれかの値を設定可能である。 The workload “product browsing” includes the AP parameter “AP processing load” in addition to the AP parameter “data size”. The AP parameter “AP processing load” is a parameter for estimating the CPU processing time of “AP processing” per transaction. In the parameter conversion formula, a high value, a medium value, or a low value can be set for the AP parameter “AP processing load”.
　ＡＰパラメータ「ＡＰ処理負荷」が「高」は、たとえば、ＣＰＵ処理時間が１００ミリ秒（以降、「ｍｓ」と表す）であることを表す。ＡＰパラメータ「ＡＰ処理負荷」が「中」は、たとえば、ＣＰＵ処理時間が５０ｍｓであることを表す。ＡＰパラメータ「ＡＰ処理負荷」が「中」は、たとえば、ＣＰＵ処理時間が５ｍｓであることを表す。 The AP parameter “AP processing load” “high” indicates, for example, that the CPU processing time is 100 milliseconds (hereinafter referred to as “ms”). When the AP parameter “AP processing load” is “medium”, for example, the CPU processing time is 50 ms. When the AP parameter “AP processing load” is “medium”, for example, the CPU processing time is 5 ms.
　この場合、ユーザは、「ＡＰ処理負荷」は高、中、または、低のいずれが適切かをあらかじめ判定する。この方法は、ＣＰＵ処理時間を明示的に見積もる場合に比べると容易である。 In this case, the user determines in advance whether the “AP processing load” is high, medium, or low. This method is easier than when the CPU processing time is explicitly estimated.
　ワークロード「購入処理」におけるパラメータ変換式は、ＡＰ機能における１トランザクションあたりのサービス時間が、ＡＰパラメータ「ＡＰ処理時間」と等しいことを表す。この場合、ＡＰパラメータ「ＡＰ処理時間」を上流ＳＥは、ＡＰ機能の１トランザクション当たりのＣＰＵ処理時間を明示的に見積もることにより入力する。 The parameter conversion formula in the workload “purchase processing” indicates that the service time per transaction in the AP function is equal to the AP parameter “AP processing time”. In this case, the upstream SE inputs the AP parameter “AP processing time” by explicitly estimating the CPU processing time per transaction of the AP function.
　ＡＰ固有処理は、あるワークロードにおいて、機能が行う処理を表す。 AP specific processing represents processing performed by a function in a certain workload.
　図４は、ワークロードが「購入処理」である場合において、「ＡＰ機能」におけるＡＰ固有処理を表すシーケンス図である。図４に示すように、ＡＰサーバは、データベース（以降、「ＤＢ」と表す。）への問い合わせを同期的に実行しながら、作業スレッドを生成し、生成した作業スレッドにおいて並行してあるいは、擬似的に並行して（以降、単に、「並列に」と表す。）作業する。ここで、図４中のメッセージ「ＡＰ処理」、「ＤＢ問い合わせ」、「結果」、及び、「ＤＢ結果」は、それぞれ、図２Ａ及び図２Ｂに示す機能フロー図中のメッセージ「ＡＰ処理」、「ＤＢ問い合わせ」、「結果」、「ＤＢ結果」に関連付けされている。 FIG. 4 is a sequence diagram showing AP-specific processing in the “AP function” when the workload is “purchase processing”. As shown in FIG. 4, the AP server generates a work thread while synchronously executing an inquiry to a database (hereinafter referred to as “DB”), and in parallel with the generated work thread, Work in parallel (hereinafter simply referred to as “parallel”). Here, the messages “AP processing”, “DB inquiry”, “result”, and “DB result” in FIG. 4 are the messages “AP processing” in the functional flow diagrams shown in FIGS. 2A and 2B, respectively. It is associated with “DB inquiry”, “result”, and “DB result”.
　なお、本実施形態においては、ＡＰ固有処理を、シーケンス図を用いて記述するが、コラボレーション図やアクティビティ図を用いて記述してもよい。 In the present embodiment, the AP-specific process is described using a sequence diagram, but may be described using a collaboration diagram or an activity diagram.
　次に、性能見積もり装置１０における動作について説明する。 Next, the operation of the performance estimation apparatus 10 will be described.
　まず、本実施形態におけるユースケース（Ｕｓｅ　Ｃａｓｅ）について図５Ａ乃至図５Ｃを用いて説明する。図５Ａ乃至図５Ｃは、本発明の実施の形態におけるユースケースの一例を表す図である。 First, a use case (Use Case) in the present embodiment will be described with reference to FIGS. 5A to 5C. 5A to 5C are diagrams illustrating an example of a use case in the embodiment of the present invention.
　図５Ａ乃至図５Ｃに示すように、後述の説明においては、（ａ）乃至（ｃ）の３種類のアクタを想定する。 As shown in FIGS. 5A to 5C, three types of actors (a) to (c) are assumed in the following description.
　　（ａ）システムモデルを作成し、作成したシステムモデルをリポジトリに登録するプラットフォームアーキテクト、
　　（ｂ）ＡＰモデルを作成して、作成したＡＰモデルをリポジトリに登録するアプリケーション開発者、
　　（ｃ）リポジトリを参照することによりモデルを検索し、検出したモデルに基づいてアプリケーションの性能を見積もり、見積もった性能を顧客に対するシステム提案に活用する上流ＳＥ。 (A) Platform architect who creates a system model and registers the created system model in the repository,
(B) An application developer who creates an AP model and registers the created AP model in the repository;
(C) An upstream SE that searches a model by referring to a repository, estimates the performance of an application based on the detected model, and uses the estimated performance for system proposal to a customer.
　プラットフォームアーキテクトは、機能レベルで同一と見なせるシステムモデルに対して、同一の機能フロー図を持つシステムモデルを作成する。図５Ａ乃至図５Ｃが示す例では、プラットフォームアーキテクトは、同一の機能フロー図を持つ異なる２つのシステムモデル「Ｗｅｂ３層システムモデル（ＯＳＳ）」と「Ｗｅｂ３層システムモデル（高可用）」とを作成する。 The platform architect creates a system model with the same functional flow diagram for the system model that can be considered the same at the functional level. In the example shown in FIGS. 5A to 5C, the platform architect creates two different system models “Web 3-tier system model (OSS)” and “Web 3-tier system model (high availability)” having the same functional flow diagram. To do.
　この運用を遵守すれば、アプリケーション開発者は、機能フロー図以外のシステムモデルの詳細を考慮することなく、ＡＰモデルを作成できる。さらに、プラットフォームアーキテクトは、アプリケーションのことを考慮することなく、システムモデルを蓄積することができる。 If this operation is observed, an application developer can create an AP model without considering the details of the system model other than the function flow diagram. Furthermore, the platform architect can accumulate system models without considering the application.
　上流ＳＥは、ＡＰモデルとシステムモデルを選択し、選択したＡＰモデルが含むそれぞれのワークロードに関するＡＰパラメータを設定する。これにより、上流ＳＥは、アプリケーションやプラットフォーム等の詳細を知らなくても、ＩＴシステムが有する性能を見積もることができる。 The upstream SE selects an AP model and a system model, and sets AP parameters related to each workload included in the selected AP model. Thereby, the upstream SE can estimate the performance of the IT system without knowing the details of the application and the platform.
　次に、システムの性能見積もり処理の流れについて説明する。 Next, the flow of the system performance estimation process will be described.
　まず、上流ＳＥは、モデル選択部１０１を介して、顧客に提案するアプリケーションを選択する。その後、モデル選択部１０１は、そのアプリケーションを実現可能なシステムモデルの一覧を表示する。具体的には、モデル選択部１０１は、ＡＰモデルに関連付けされた処理フローを全て含むシステムモデルを検索することによって、システムモデルの一覧を表示する。 First, the upstream SE selects an application to be proposed to the customer via the model selection unit 101. Thereafter, the model selection unit 101 displays a list of system models that can realize the application. Specifically, the model selection unit 101 displays a list of system models by searching for system models that include all processing flows associated with the AP model.
　次に、上流ＳＥは、ＡＰパラメータ入力部１０２を介して、案件固有の事情に応じてＡＰパラメータを入力する。その後、モデル合成部１０３は、上流ＳＥが入力したＡＰパラメータと、選択されたシステムモデルと、ＡＰモデルとに基づいて、ＩＴシステムが有する性能を見積もる場合に参照するモデル（以降、「性能モデル」と表す。）を生成する。 Next, the upstream SE inputs AP parameters according to the circumstances specific to the case via the AP parameter input unit 102. Thereafter, the model combining unit 103 refers to a model (hereinafter referred to as “performance model”) that is referred to when estimating the performance of the IT system based on the AP parameter input by the upstream SE, the selected system model, and the AP model. Is generated.
　次に、見積もり部１０４は、所定の算出手順に従い、モデル合成部１０３が生成した性能モデルに基づいて、上流ＳＥが入力したパラメータにおける性能を見積った結果を、出力装置１０６に表示する。見積もり部１０４は、所定の算出手順として、性能モデルを解析的に計算することにより見積もり値を算出してもよいし、シミュレーションすることにより見積もり値を算出してもよい。たとえば、特願２０１１－１３２３７７号（以降、「特許文献２」と表す。）が開示する手法を用いて、見積もり部１０４を実現してもよい。 Next, the estimation unit 104 displays, on the output device 106, the result of estimating the performance of the parameters input by the upstream SE based on the performance model generated by the model synthesis unit 103 according to a predetermined calculation procedure. As the predetermined calculation procedure, the estimation unit 104 may calculate an estimated value by analytically calculating a performance model, or may calculate an estimated value by simulation. For example, the estimation unit 104 may be realized using a technique disclosed in Japanese Patent Application No. 2011-132377 (hereinafter referred to as “Patent Document 2”).
　次に、図６を参照しながら、モデル合成部１０３における処理について説明する。 Next, processing in the model synthesis unit 103 will be described with reference to FIG.
　ここでは、上流ＳＥが、図２のシステムモデルと図３のＡＰモデルとが選択する場合において、モデル合成部１０３がワークロード「購入処理」に関連する性能モデルを生成する例を参照しながら説明する。 Here, when the upstream SE selects the system model of FIG. 2 and the AP model of FIG. 3, the model composition unit 103 will be described with reference to an example of generating a performance model related to the workload “purchase processing”. To do.
　まず、対象とするワークロードに関連付けされた処理フローとＡＰ固有処理とに基づいて、機能レベルにおける実行フローを生成する（ステップＳ１）。図７は、変換後の実行フローをアクティビティ図で表す場合に、機能レベルにおける実行フローを生成する様子を表す図である。 First, an execution flow at the function level is generated based on the processing flow associated with the target workload and the AP specific processing (step S1). FIG. 7 is a diagram illustrating a state in which an execution flow at the function level is generated when the converted execution flow is represented by an activity diagram.
　本実施形態においては、処理フローを、コラボレーション図を用いて記述するとともに、ＡＰ固有処理を、シーケンス図を用いて記述する。両者は、メッセージ名を介して関連付けされている。機能レベルにおける実行フローを生成する処理は、生成手順１乃至生成手順３に従い実現できる。 In this embodiment, the process flow is described using a collaboration diagram, and the AP-specific process is described using a sequence diagram. Both are associated via a message name. The process of generating the execution flow at the function level can be realized according to generation procedure 1 to generation procedure 3.
　　（生成手順１）処理フロー中のメッセージの順に従い、メッセージをノード（すなわち、グラフ理論における節点）に関連付けすることにより、実行フロー（以降、「第１実行フロー」と表す。）を生成する、
　　（生成手順２）ＡＰ固有処理中の処理内容を、メッセージをノードに関連付けすることにより、実行フロー（以降、「第２実行フロー」と表す。）を生成する。シーケンス図に関する内容を、たとえば、参考文献２が記述する方法を用いて解釈し、解釈したシーケンス図に関する内容と同等のアクティビティ図を用いて表してもよい、
　　（生成手順３）第１実行フローと、第２実行フローとを合成する。第１実行フロー中のノード、及び、第２実行フロー中のノードは、メッセージ名に関連付けされている。そこで、合成処理においては、第１実行フロー及び第２実行フローにおいて、同じメッセージ名に関連付けされたノードを同一のノードと見なす。 (Generation Procedure 1) An execution flow (hereinafter referred to as “first execution flow”) is generated by associating a message with a node (that is, a node in the graph theory) in the order of messages in the processing flow.
(Generation Procedure 2) An execution flow (hereinafter referred to as “second execution flow”) is generated by associating a message with a node for the processing content in the AP-specific processing. The content related to the sequence diagram may be interpreted using, for example, the method described in Reference 2, and may be represented using an activity diagram equivalent to the content related to the interpreted sequence diagram.
(Generation Procedure 3) The first execution flow and the second execution flow are synthesized. The node in the first execution flow and the node in the second execution flow are associated with the message name. Therefore, in the synthesis process, the nodes associated with the same message name are regarded as the same node in the first execution flow and the second execution flow.
　（参考文献２）Ｚｏｌｔａｎ　Ｍｉｃｓｋｅｉ，　Ｈｅｌｅｎｅ　Ｗａｅｓｅｌｙｎｃｋ，　“Ｔｈｅ　ｍａｎｙ　ｍｅａｎｉｎｇｓ　ｏｆ　ＵＭＬ２　Ｓｅｑｕｅｎｃｅ　Ｄｉａｇｒａｍｓ：　ａ　ｓｕｒｖｅｙ“，　Ｓｏｆｔｗａｒｅ　ａｎｄ　Ｓｙｓｔｅｍｓ　Ｍｏｄｅｌｉｎｇ　Ｖｏｌｕｍｅ　１０，　Ｎｕｍｂｅｒ　４，　４８９－５１４。 (Reference 2) Zoltan Mickei, Helene Weselynck, “The many means of UML2 Sequence Diagrams: a survey”, Software Quantum 9: 48
　次に、ＡＰパラメータ入力部１０２に入力されたＡＰパラメータと、該ＡＰパラメータに関連付けされたパラメータ変換式とに基づいて、機能レベルにおけるパラメータを算出する（ステップＳ２）。 Next, parameters at the function level are calculated based on the AP parameters input to the AP parameter input unit 102 and the parameter conversion formula associated with the AP parameters (step S2).
　たとえば、入力されたＡＰパラメータが、以下のように表されるとする。 Suppose, for example, that the input AP parameter is expressed as follows.
　　データサイズ＝１メガバイト（以降、「ＭＢ」と表す）、
　　ＡＰ処理時間＝０．０１秒（以降、「ｓ」と表す）、
　　背景処理時間＝０．００５（ｓ）。 Data size = 1 megabyte (hereinafter referred to as “MB”),
AP processing time = 0.01 seconds (hereinafter referred to as “s”),
Background processing time = 0.005 (s).
　パラメータ変換式に、入力されたＡＰパラメータの値を代入すると、以下の値が得られる。代入により得られた値が、機能フロー図中のパラメータを表す。 ∙ Substituting the input AP parameter value into the parameter conversion formula gives the following values. The value obtained by the substitution represents the parameter in the function flow diagram.
　　動的処理．ｄｅｍａｎｄ＝０．０１、
　　ＡＰ処理．ｄｅｍａｎｄ＝０．０１、
　　背景処理．ｄｅｍａｎｄ＝０．００５。 Dynamic processing. demand = 0.01,
AP processing. demand = 0.01,
Background processing. demand = 0.005.
　次に、機能レベルにおける実行フローを、プロセスレベルにおける実行フローに変換する（ステップＳ３）。本実施形態においては、機能とプロセスとは、一対一に関連付けされる。このため、この変換処理は、機能レベルにおける実行フロー中のノードを、該機能に関連付けされたプロセスを表すノードに置き換えることにより実現できる。なお、機能とプロセスとが多対一に関連付けされる場合であっても、一対一に関連付けされると同様に処理できる。 Next, the execution flow at the function level is converted into an execution flow at the process level (step S3). In the present embodiment, functions and processes are associated one to one. For this reason, this conversion processing can be realized by replacing a node in the execution flow at the function level with a node representing a process associated with the function. Note that even when functions and processes are associated in a many-to-one relationship, processing can be performed in a similar manner when associated in a one-to-one relationship.
　次に、プロセスレベルにおける実行フローを性能モデルに変換する（ステップＳ４）。この処理の例は、たとえば、特許文献２に記載されている。このため、本実施形態においては、変換する処理に関する詳細な説明を省略する。 Next, the execution flow at the process level is converted into a performance model (step S4). An example of this processing is described in Patent Document 2, for example. For this reason, in this embodiment, the detailed description regarding the process to convert is abbreviate | omitted.
　上述した処理を行うことにより、本実施形態により、以下の効果を得ることができる。 By performing the above-described processing, the following effects can be obtained according to the present embodiment.
　まず、上流ＳＥは、大幅に容易にパラメータを入力することができる。これは、プラットフォームに関する情報が上流ＳＥに隠蔽されるため、上流ＳＥは、ＡＰパラメータのみを設定すれば済むためである。 First, the upstream SE can input parameters significantly easily. This is because the information about the platform is hidden in the upstream SE, and the upstream SE only needs to set the AP parameter.
　また、システムモデルを再利用することが大幅に容易になる。これは、図４に示すようなＡＰ固有処理がシステムモデルに記載されないために、システムモデルに従いアプリケーションを実装する場合と、固有の処理とが分離されるためである。 Also, it becomes much easier to reuse the system model. This is because the AP-specific processing as shown in FIG. 4 is not described in the system model, so that the case where the application is implemented according to the system model and the specific processing are separated.
　なお、本実施形態においては、ＡＰモデルとシステムモデルとを最初に決定するが、ＡＰモデルの決定及びＡＰパラメータの入力を先に行い、その後、システムモデルを選択するようにしてもよい。これにより、見積もりの結果、提案するＩＴシステムがユーザと価格や性能に関する面において折り合わなかった場合に、ＡＰモデルとＡＰパラメータの入力結果とを変更することなく、システムモデルのみ変更することにより再見積もりをすることが容易になる。 In this embodiment, the AP model and the system model are determined first. However, the AP model may be determined and the AP parameter may be input first, and then the system model may be selected. As a result, if the proposed IT system does not agree with the user in terms of price and performance as a result of estimation, re-estimation can be performed by changing only the system model without changing the AP model and AP parameter input results. It becomes easy to do.
　また、本実施形態においては、システムモデルが含む機能フロー図と、ＡＰモデルが含むＡＰ固有処理とを合成する。しかし、システムモデル中に処理のシーケンスを記述せずに、ＡＰモデル中に処理のシーケンス（機能フロー）を記述してもよい。 In the present embodiment, the function flow diagram included in the system model and the AP specific process included in the AP model are combined. However, the processing sequence (functional flow) may be described in the AP model without describing the processing sequence in the system model.
　たとえば、システムモデルを図８に示すように記述し、ＡＰモデルを図９Ａに示すようように記述し、これらを含まず、機能を定義する機能図を持つシステムモデルの例である。機能とプロセスとの関係を、アロケーションを用いて記述する。図９Ａにおいては、ＡＰ固有処理の代わりに、そのワークロード中の機能レベルにおける実行シーケンスが記述されている。このような構成にした場合、ＡＰモデル作成者が、より詳細にＡＰの処理をモデル化しやすい。 For example, a system model is described as shown in FIG. 8 and an AP model is described as shown in FIG. 9A, which is an example of a system model that does not include these and has a function diagram that defines functions. Describe the relationship between functions and processes using allocation. In FIG. 9A, instead of AP-specific processing, an execution sequence at a function level in the workload is described. In such a configuration, the AP model creator can easily model the AP processing in more detail.
　たとえば、図９Ａにおいて、「ＨＰ表示」と、「データサイズ」と、「静的閲覧ｄｅｍａｎｄ＝データサイズ×０．０１」と、「機能Ｂ」とは、関連付けされている。 For example, in FIG. 9A, “HP display”, “data size”, “static browsing demand = data size × 0.01”, and “function B” are associated with each other.
　図９Ａにおける、機能Ｂ、機能Ｃ及び機能Ｄは、それぞれ、図９Ｂにおけるシーケンス図、図９Ｃにおけるシーケンス図、及び、図９Ｄにおけるシーケンス図を表す。 9A, function B, function C, and function D represent the sequence diagram in FIG. 9B, the sequence diagram in FIG. 9C, and the sequence diagram in FIG. 9D, respectively.
　たとえば、図９Ｂにおいて、クライアント機能は、ＷＷＷ機能に静的閲覧を要求する。次に、ＷＷＷ機能は、該要求に応じて処理を行い、処理した結果をクライアント機能に送信する。 For example, in FIG. 9B, the client function requests static browsing to the WWW function. Next, the WWW function performs processing in response to the request, and transmits the processing result to the client function.
　すなわち、本実施形態に係る性能見積もり装置によれば、システム設計が容易になる。 That is, the performance estimation apparatus according to the present embodiment facilitates system design.
　以上、上述した実施形態を模範的な例として本発明を説明した。しかし、本発明は、上述した実施形態には限定されない。即ち、本発明は、本発明のスコープ内において、当業者が理解し得る様々な態様を適用することができる。 The present invention has been described above using the above-described embodiment as an exemplary example. However, the present invention is not limited to the above-described embodiment. That is, the present invention can apply various modes that can be understood by those skilled in the art within the scope of the present invention.
　１０　　性能見積もり装置
　１０１　　モデル選択部
　１０２　　ＡＰパラメータ入力部
　１０３　　モデル合成部
　１０４　　見積もり部
　１０５　　モデル記憶部
　１０６　　出力装置
　２０　　計算処理装置
　２１　　ＣＰＵ
　２２　　メモリ
　２３　　ディスク
　２４　　不揮発性記録媒体
　２５　　入力装置
　２６　　出力装置 DESCRIPTION OF SYMBOLS 10 Performance estimation apparatus 101 Model selection part 102 AP parameter input part 103 Model synthetic | combination part 104 Estimation part 105 Model memory | storage part 106 Output apparatus 20 Computation processing apparatus 21 CPU
22 Memory 23 Disk 24 Non-volatile recording medium 25 Input device 26 Output device

Claims (6)

  1. 　ＩＴシステムが有するプラットフォームに関する情報を含むシステムモデルと、前記ＩＴシステムにおける機能を実現するアプリケーションに固有な処理であるＡＰ固有処理に関する情報及び前記アプリケーションに含まれるパラメータに関する情報を含むＡＰモデルとが、共通して含むメッセージに基づき、前記システムモデルと前記ＡＰモデルとを合成することにより、性能モデルを作成するモデル合成手段と、
    　前記性能モデルに基づき、所定の算出手順に従い、前記ＩＴシステムにおける性能を算出する見積り手段と
    　を備える性能見積もり装置。     A system model including information related to a platform included in the IT system, and an AP model including information related to AP-specific processing that is processing specific to an application that realizes the function in the IT system and information related to parameters included in the application are common. Model synthesis means for creating a performance model by synthesizing the system model and the AP model based on the message
    A performance estimation apparatus comprising: an estimation unit that calculates performance in the IT system according to a predetermined calculation procedure based on the performance model.
  2. 　前記システムモデルは、前記プラットフォームが有する構成を抽象化した機能レベルにおける前記ＩＴシステムに関する処理を表す機能フローを含み、
    　前記ＡＰモデルは、前記ＡＰ固有処理を表すシーケンスを含み、
    　前記モデル合成手段は、前記ＡＰモデルと前記システムモデルを合成することにより、機能レベルにおける実行フローを生成する
    　請求項１に記載の性能見積もり装置。     The system model includes a function flow representing processing related to the IT system at a function level that abstracts the configuration of the platform,
    The AP model includes a sequence representing the AP specific process;
    The performance estimation apparatus according to claim 1, wherein the model synthesis unit generates an execution flow at a function level by synthesizing the AP model and the system model.
  3. 　前記システムモデルは、前記プラットフォームが有する構成を抽象化した一連の機能を表す機能フローを含み、
    　前記ＡＰモデルは、前記アプリケーションが有する固有の機能間における処理に関するシーケンスを表すＡＰ機能フローを含み、
    　前記モデル合成手段は、前記ＡＰモデルと前記システムモデルを合成することにより、機能レベルにおける実行フローを生成する
    　請求項１に記載の性能見積もり装置。     The system model includes a functional flow representing a series of functions abstracted from the configuration of the platform,
    The AP model includes an AP function flow representing a sequence related to processing between unique functions of the application,
    The performance estimation apparatus according to claim 1, wherein the model synthesis unit generates an execution flow at a function level by synthesizing the AP model and the system model.
  4. 　前記ＡＰモデルは、アプリケーションレベルにおいて設計する項目を表すＡＰパラメータと、前記ＡＰパラメータが表す内容を前記システムモデルが含むパラメータに変換するパラメータ変換式を含む、
    　請求項２または請求項３に記載の性能見積もり装置。     The AP model includes an AP parameter representing an item designed at an application level, and a parameter conversion formula for converting the content represented by the AP parameter into a parameter included in the system model.
    The performance estimation apparatus according to claim 2 or claim 3.
  5. 　情報処理装置において、ＩＴシステムが有するプラットフォームに関する情報を含むシステムモデルと、前記ＩＴシステムにおける機能を実現するアプリケーションに固有な処理であるＡＰ固有処理に関する情報及び前記アプリケーションに含まれるパラメータに関する情報を含むＡＰモデルとが、共通して含むメッセージに基づき、前記システムモデルと前記ＡＰモデルとを合成することにより、性能モデルを作成し、前記性能モデルに基づき、所定の算出手順に従い、前記ＩＴシステムにおける性能を算出する性能見積もり方法。 In the information processing apparatus, an AP that includes a system model including information related to a platform included in the IT system, information related to an AP specific process that is a process specific to an application that realizes a function in the IT system, and information related to a parameter included in the application Based on the message that the model includes in common, the system model and the AP model are combined to create a performance model. Based on the performance model, the performance in the IT system is determined according to a predetermined calculation procedure. Performance estimation method to be calculated.
  6. 　ＩＴシステムが有するプラットフォームに関する情報を含むシステムモデルと、前記ＩＴシステムにおける機能を実現するアプリケーションに固有な処理であるＡＰ固有処理に関する情報及び前記アプリケーションに含まれるパラメータに関する情報を含むＡＰモデルとが、共通して含むメッセージに基づき、前記システムモデルと前記ＡＰモデルとを合成することにより、性能モデルを作成するモデル合成機能と、
    　前記性能モデルに基づき、所定の算出手順に従い、前記ＩＴシステムにおける性能を算出する見積り機能と
    　をコンピュータに実現させる性能見積もりプログラム。     A system model including information related to a platform included in the IT system, and an AP model including information related to AP-specific processing that is processing specific to an application that realizes the function in the IT system and information related to parameters included in the application are common. A model synthesis function for creating a performance model by synthesizing the system model and the AP model based on the message
    A performance estimation program for causing a computer to realize an estimation function for calculating performance in the IT system according to a predetermined calculation procedure based on the performance model.
PCT/JP2013/006436 2012-11-02 2013-10-30 Performance estimating system, perform ance estimating method, and program WO2014068979A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2014544321A JPWO2014068979A1 (en) 2012-11-02 2013-10-30 Performance estimation device, performance estimation method, and program

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2012-242606 2012-11-02
JP2012242606 2012-11-02

Publications (1)

Publication Number Publication Date
WO2014068979A1 true WO2014068979A1 (en) 2014-05-08

Family

ID=50626927

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2013/006436 WO2014068979A1 (en) 2012-11-02 2013-10-30 Performance estimating system, perform ance estimating method, and program

Country Status (2)

Country Link
JP (1) JPWO2014068979A1 (en)
WO (1) WO2014068979A1 (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002183416A (en) * 2000-12-15 2002-06-28 Hitachi Ltd System proposing method, device for performing it, storage medium storing its processing program
JP2011145937A (en) * 2010-01-15 2011-07-28 Nec Corp System, method and program for supporting program inspection
JP2012146015A (en) * 2011-01-07 2012-08-02 Nec Corp Performance evaluation system, performance evaluation method and performance evaluation program

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002183416A (en) * 2000-12-15 2002-06-28 Hitachi Ltd System proposing method, device for performing it, storage medium storing its processing program
JP2011145937A (en) * 2010-01-15 2011-07-28 Nec Corp System, method and program for supporting program inspection
JP2012146015A (en) * 2011-01-07 2012-08-02 Nec Corp Performance evaluation system, performance evaluation method and performance evaluation program

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
AKIRA KADO: "Throughput Performance Verification of Airship Autopilot Software : A Case Study for Evaluating Performance Verification of Software Design Model", IEICE TECHNICAL REPORT, vol. 112, 3 May 2012 (2012-05-03), pages 25 - 30 *
TSUTOMU SHU: "UML o Mochiita System Level Sekkei Shuho no Teian", DA SYMPOSIUM, 2002, vol. 2002, 24 July 2002 (2002-07-24), pages 49 - 54 *

Also Published As

Publication number Publication date
JPWO2014068979A1 (en) 2016-09-08

Similar Documents

Publication Publication Date Title
US8140319B2 (en) Method and system for predicting system performance and capacity using software module performance statistics
TW201723886A (en) Information-processing equipment based on a spreadsheet
Verdickt et al. Automatic inclusion of middleware performance attributes into architectural UML software models
JP2017519314A (en) A computing platform that utilizes models for integrated design of products, materials and manufacturing processes
Cito et al. Runtime metric meets developer: building better cloud applications using feedback
McKee et al. The internet of simulation, a specialisation of the internet of things with simulation and workflow as a service (sim/wfaas)
Lenhard et al. Lessons learned from evaluating workflow management systems
JP4872702B2 (en) Distributed workflow simulation system, method, and program
JP2014174609A (en) Hardware configuration estimation system, hardware configuration estimation method and hardware configuration estimation program
Walter et al. Tools for declarative performance engineering
WO2014068979A1 (en) Performance estimating system, perform ance estimating method, and program
Lehrig et al. Performance prototyping with protocom in a virtualised environment: A case study
Kano et al. Analysis and simulation of business solutions in a service-oriented architecture
Westermann et al. Towards performance prediction of large enterprise applications based on systematic measurements
Weber et al. Online performance analysis with the Vampir tool set
Alhaj et al. Using aspects for platform-independent to platform-dependent model transformations
Qureshi et al. On integrating east-adl and uppaal for embedded system architecture verification
JP2012150740A (en) Program generating device and method
JPWO2019021898A1 (en) Development support equipment
Hennig et al. Performance prototyping–generating and simulating a distributed IT-system from UML models
Marek Generating Petri Nets for Reliability Analysis of Smart Grids From UML Diagrams
Truong et al. Dipas: A distributed performance analysis service for grid service-based workflows
Dalman et al. Service workflows and distributed computing methods for 13C metabolic flux analysis
Mazurkiewicz Dependability metrics for network systems—analytical and experimental analysis
Kumar A Novel Testing Model for SOA based Services.

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 13851453

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2014544321

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 13851453

Country of ref document: EP

Kind code of ref document: A1