軟件體系結(jié)構(gòu)體系結(jié)構(gòu)評審ArhitecturalReviews

《軟件體系結(jié)構(gòu)體系結(jié)構(gòu)評審ArhitecturalReviews》由會員分享，可在線閱讀，更多相關(guān)《軟件體系結(jié)構(gòu)體系結(jié)構(gòu)評審ArhitecturalReviews（49頁珍藏版）》請在裝配圖網(wǎng)上搜索。

1、軟件體系結(jié)構(gòu)軟件體系結(jié)構(gòu)體系結(jié)構(gòu)評審體系結(jié)構(gòu)評審Architectural ReviewsArchitectural Reviews孫志崗 2 22021-12-19Architectural ReviewsnArchitectural reviews are a key step in architecture-based development. They include a method (such as SAAM or ATAM), but the full practice includes more than that.體系結(jié)構(gòu)評審是開發(fā)過程中的一個重要環(huán)節(jié)。有專門的評體系結(jié)構(gòu)評審

2、是開發(fā)過程中的一個重要環(huán)節(jié)。有專門的評審方法審方法(SAAM、ATAM)，但在實踐中并不僅僅是這些，但在實踐中并不僅僅是這些nWe will discusscost/benefits of architectural reviewsdifferent architectural review techniquespreconditions for an architectural reviewarchitectural review activitiesoutputs from an architectural review 3 32021-12-19Costs of Architectura

3、l ReviewsnThere are three different types of costs of holding regular architectural reviews.direct staff costs直接員工開銷直接員工開銷organizational overhead組織管理開銷組織管理開銷indirect staff costs間接員工開銷間接員工開銷 4 42021-12-19Examples of Direct Staff CostsnAT&T300 full-scale reviews done on projects of 700 staff-days or l

4、onger全面評審了全面評審了300個項目，每個項目的開發(fā)時間都是個項目，每個項目的開發(fā)時間都是700人日以上人日以上average cost per review: 70 staff days每次評審的平均開銷：每次評審的平均開銷：70人日人日nRational Software30 reviews done on projects with at least 500 KSLOC each評審了評審了30個項目，每個項目最少個項目，每個項目最少500 KSLOCaverage cost per review: $50,000評審平均花費：評審平均花費：$50,000 5 52021-12-1

5、9Examples of Direct Staff CostsnSAAM evaluations12 reviews done on projects ranging from 100 KSLOC to 1,000 KSLOC12次評審，項目規(guī)模從次評審，項目規(guī)模從100 KSLOC到到1,000 KSLOCaverage cost per review: 14 to 20 staff-days平均花費：平均花費：14到到20人日人日nATAM evaluations12 reviews done on projects ranging from 250 KSLOC to 2,500 KSLO

6、C12次評審，項目規(guī)模從次評審，項目規(guī)模從250 KSLOC到到2,500 KSLOCaverage cost per review: 40 to 70 staff days平均花費：平均花費：40到到70人日人日 6 62021-12-19Example of Organizational OverheadnCosts of establishing a corporate review unit建立評審組的開銷建立評審組的開銷management overheadcommunication expensesstaffing the unit給評審組配備人手給評審組配備人手relocatin

7、g personnel to a central locationtraining 7 72021-12-19Example of Indirect Staff CostsnUsing senior designers for evaluations instead of designing讓高級設(shè)計人員參與評審而不是設(shè)計讓高級設(shè)計人員參與評審而不是設(shè)計loss of productivity (due to reassignment of superior designers)生產(chǎn)力的下降生產(chǎn)力的下降(高級設(shè)計師被委派新任務(wù)高級設(shè)計師被委派新任務(wù))time spent training st

8、aff in review techniques培訓(xùn)員工評審技術(shù)的時間消耗培訓(xùn)員工評審技術(shù)的時間消耗 8 82021-12-19Benefits of Architectural ReviewsnFive different types of benefits result from holding architectural reviews.financialforces preparation for review為評審做準(zhǔn)備也是一種推動力為評審做準(zhǔn)備也是一種推動力early detection of problems盡早地發(fā)現(xiàn)問題盡早地發(fā)現(xiàn)問題validation of requirem

9、ents確認(rèn)需求確認(rèn)需求improved architectures提高體系結(jié)構(gòu)質(zhì)量提高體系結(jié)構(gòu)質(zhì)量 9 92021-12-19Financial Benefits of ReviewsnAT&T estimated that each reviewed project saves 10% of total cost as a result of the review. Thus, a 70-staff-day review of projects of 700 staff-days pays for itself.AT&T估算，每個被評審的項目節(jié)省估算，每個被評審的項目節(jié)省10%的成本的成本

10、nConsultants who perform architecture evaluations report 80% repeat business.參與評審的顧問報告了參與評審的顧問報告了80%的重復(fù)項目的重復(fù)項目 10102021-12-19Forces Preparation for ReviewnDocumentation/specifications must be provided, hence they must exist or be created.文檔和說明書必須存在文檔和說明書必須存在nSome reviews use standard questions, and

11、the architect can prepare ahead to ensure that the architecture scores well.評審有一些標(biāo)準(zhǔn)問題。體系結(jié)構(gòu)師可以事先準(zhǔn)備，確保這評審有一些標(biāo)準(zhǔn)問題。體系結(jié)構(gòu)師可以事先準(zhǔn)備，確保這個體系結(jié)構(gòu)可能得高分個體系結(jié)構(gòu)可能得高分nReviews make the criteria for evaluation explicit by prioritizing requirements or quality goals.通過排定需求和質(zhì)量目標(biāo)的順序，評審使評估的標(biāo)準(zhǔn)變得通過排定需求和質(zhì)量目標(biāo)的順序，評審使評估的標(biāo)準(zhǔn)變得清晰清晰 11

12、112021-12-19Early Detection of ProblemsnThe problems that can be found by an architectural level inspection include在體系結(jié)構(gòu)層進(jìn)行檢查可以發(fā)現(xiàn)的問題在體系結(jié)構(gòu)層進(jìn)行檢查可以發(fā)現(xiàn)的問題unreasonable requirementsperformance problemsproblems associated with potential future modifications與未來的修改有關(guān)的問題與未來的修改有關(guān)的問題nThe earlier in the life cycl

13、e that problems are found, the easier it is to fix them. 12122021-12-19Validation of RequirementsnReviews put stakeholders in the same room with each other, often for the first time.評審使各種角色坐在同一個房間內(nèi)。通常，這是第一次評審使各種角色坐在同一個房間內(nèi)。通常，這是第一次uncovers conflicts and tradeoffs解開沖突和尋找折中解開沖突和尋找折中provides a forum for

14、 negotiated resolution of problems提供了一個論壇，來磋商問題的解決辦法提供了一個論壇，來磋商問題的解決辦法nIt often results in the generation of new requirements or the clarification of existing requirements.結(jié)果通常會是產(chǎn)生新的需求和凈化已有的需求結(jié)果通常會是產(chǎn)生新的需求和凈化已有的需求 13132021-12-19Improved ArchitecturesnDevelopment organizations anticipate types of ques

15、tions raised at reviews and開發(fā)者預(yù)期在評審時會被發(fā)現(xiàn)的問題，然后開發(fā)者預(yù)期在評審時會被發(fā)現(xiàn)的問題，然后design architectures with questions in mind在設(shè)計時，問題會時刻圍繞腦海在設(shè)計時，問題會時刻圍繞腦海prepare documentation of the type needed at review準(zhǔn)備評審時需要的文檔準(zhǔn)備評審時需要的文檔give explicit consideration to qualities to be reviewed對將被評審的質(zhì)量因素進(jìn)行更深入的考慮對將被評審的質(zhì)量因素進(jìn)行更深入的考慮 14

16、142021-12-19Review TechniquesnThere are a variety of techniques for performing architectural reviews; each has a different cost and provides different information.有很多種評審技術(shù)。每種需要不同的開銷，提供不同的信有很多種評審技術(shù)。每種需要不同的開銷，提供不同的信息息nThese techniques fall into one of two categories.questioning techniques: applied to

17、evaluate any aspect of an architecture for any given reason提問技術(shù)提問技術(shù)：用來評價體系結(jié)構(gòu)的任何方面：用來評價體系結(jié)構(gòu)的任何方面measuring techniques: applied to answer questions about a specific quality度量技術(shù)度量技術(shù)：回答關(guān)于某項質(zhì)量的問題：回答關(guān)于某項質(zhì)量的問題 15152021-12-19Questioning TechniquesnScenario-based techniquesdescribes a specific interaction bet

18、ween stakeholders and a system描述角色與系統(tǒng)交互的過程描述角色與系統(tǒng)交互的過程example: SAAM, ATAM (partially)nQuestionnaire-based techniques（基于問卷的技術(shù)）（基于問卷的技術(shù)）Some questions apply to all architectures (especially those in a single domain).有些問題被用于所有的體系結(jié)構(gòu)，尤其是在單一領(lǐng)域內(nèi)有些問題被用于所有的體系結(jié)構(gòu)，尤其是在單一領(lǐng)域內(nèi)Some questions ask about details of a

19、specific architecture.有些問題詢問一個特定結(jié)構(gòu)的細(xì)節(jié)有些問題詢問一個特定結(jié)構(gòu)的細(xì)節(jié)Some ask about the development process.n“Is there a single architect?”n“How do you ensure conformance?”“你們怎樣保證一致性？你們怎樣保證一致性？” 16162021-12-19Questioning TechniquesnQuestionnaires are reused; they reflect more maturity or experience within a domain.問

20、卷可以重用；它們體現(xiàn)了在一個領(lǐng)域內(nèi)問卷可以重用；它們體現(xiàn)了在一個領(lǐng)域內(nèi)的成熟度和經(jīng)驗是否豐富的成熟度和經(jīng)驗是否豐富nScenarios must be developed anew for each system.場景必須為每個系統(tǒng)重新開發(fā)場景必須為每個系統(tǒng)重新開發(fā) 17172021-12-19Measuring TechniquesnMetrics: quantitative interpretations of observable measures量度：可測質(zhì)量的定量說明量度：可測質(zhì)量的定量說明complexity metrics identify areas where modific

21、ations may be difficult復(fù)雜度的測量指明在哪個區(qū)域修改會很困難復(fù)雜度的測量指明在哪個區(qū)域修改會很困難performance metrics suggest areas where bottlenecks may occur性能的測量提示瓶頸會在哪里發(fā)生性能的測量提示瓶頸會在哪里發(fā)生nReviews tend to focus onresults of metrics度量的結(jié)果度量的結(jié)果assumptions underlying the interpretation of the metrics (e.g., assumed event distribution)對度量的

22、解釋中的潛在假定對度量的解釋中的潛在假定 18182021-12-19Measuring TechniquesnSimulations, prototypes, experiments: domain specific models of an architecture or performance model仿真、原型、實驗：建立體系模型或者性能模型仿真、原型、實驗：建立體系模型或者性能模型are expensive to createoften exist as part of development anywaynMay answer issues raised by questioni

23、ng technique 可以回答提問技術(shù)提出的問題可以回答提問技術(shù)提出的問題“What evidence do you have that performance is adequate?”“你憑什么說可以達(dá)到這個性能？你憑什么說可以達(dá)到這個性能？” 19192021-12-19Properties of Review Techniques評審技術(shù)評審技術(shù)通用性通用性細(xì)節(jié)等級細(xì)節(jié)等級階段階段被評審對象被評審對象提問提問技術(shù)技術(shù)問卷問卷通用通用粗糙粗糙早期早期設(shè)計和過程設(shè)計和過程場景場景特定系統(tǒng)特定系統(tǒng)中等中等中期中期設(shè)計設(shè)計度量度量技術(shù)技術(shù)度量度量通用或者特通用或者特定領(lǐng)域定領(lǐng)域好好中期中期

24、設(shè)計設(shè)計仿真仿真原型原型實驗實驗特定領(lǐng)域特定領(lǐng)域不定不定早期早期設(shè)計設(shè)計 20202021-12-19Review ProcessnAn architectural review has three major elements.preconditions: the set of necessary assets and conditions that must be in place before a review can be performed successfully前提前提：評審成功進(jìn)行之前，必要的資源和條件必須到：評審成功進(jìn)行之前，必要的資源和條件必須到位位activities o

25、f the review itself活動活動：評審本身：評審本身outputs: the report of review results輸出輸出：評審結(jié)果報告：評審結(jié)果報告 21212021-12-19Review PreconditionsnUnderstand the context of the review理解評審的上下文理解評審的上下文nAssemble the right people集合正確的人集合正確的人nSet organizational expectations and support期望和支持期望和支持nPrepare for review為評審做準(zhǔn)備為評審做準(zhǔn)備n

26、Obtain representation of the architecture獲得體系結(jié)構(gòu)文檔獲得體系結(jié)構(gòu)文檔 22222021-12-19Preconditions: ContextnPlanned reviewpart of normal development activity正常開發(fā)活動的一部分正常開發(fā)活動的一部分non-adversarial（非敵對性的）（非敵對性的）scheduled well in advance（事先安排好日程）（事先安排好日程）built into projects schedule and budget包含在項目日程和預(yù)算之內(nèi)包含在項目日程和預(yù)算之內(nèi)n

27、Unplannedusually occurs when project is in trouble通常在項目出現(xiàn)問題時采用通常在項目出現(xiàn)問題時采用often devolves into finger-pointing導(dǎo)致互相責(zé)備導(dǎo)致互相責(zé)備can be painful for project already struggling對于已經(jīng)苦苦掙扎了很久的項目而言，有些痛苦對于已經(jīng)苦苦掙扎了很久的項目而言，有些痛苦 23232021-12-19Preconditions: ContextnTiming of the reviewearly “architecture discovery revi

28、ew”nis done after requirements are set, but before the architecture is firm需求分析之后，體系結(jié)構(gòu)還沒有確定之前進(jìn)行需求分析之后，體系結(jié)構(gòu)還沒有確定之前進(jìn)行nis used to understand implications of requirements on architecture用來理解需求在體系結(jié)構(gòu)方面的隱含內(nèi)容用來理解需求在體系結(jié)構(gòu)方面的隱含內(nèi)容nchecks for requirements feasibility（檢查需求的可行性）（檢查需求的可行性）nprioritizes architectural

29、 goals（為質(zhì)量目標(biāo)排序）（為質(zhì)量目標(biāo)排序）full architectural reviewnis done when architectural documentation is available體系結(jié)構(gòu)文檔可用時進(jìn)行體系結(jié)構(gòu)文檔可用時進(jìn)行nis used to evaluate qualities of proposed architecture評價被評體系結(jié)構(gòu)的質(zhì)量評價被評體系結(jié)構(gòu)的質(zhì)量 24242021-12-19Preconditions: PeoplenRepresentatives of project（項目代表）（項目代表）architectdesigners of m

30、ajor modulesstakeholders representativesnReview team members (ideally, a separate unit)評審團(tuán)成員評審團(tuán)成員(理想情況下是一個獨立組織理想情況下是一個獨立組織)at least one domain expert included至少有一個領(lǐng)域?qū)＜抑辽儆幸粋€領(lǐng)域?qū)＜襰omeone to handle logistics (e.g., meeting rooms, supplies)后勤人員后勤人員(會議室、茶水、食品會議室、茶水、食品)librarian to organize documentation文檔

31、管理員文檔管理員“apprentice reviewer” for growth成長中的評審員學(xué)徒成長中的評審員學(xué)徒 25252021-12-19Preconditions: Review TeamnThe review team must be assembled so thatmembers are perceived as impartial, objective, and respected成員要公正、客觀和被尊重成員要公正、客觀和被尊重members devote their full-time effort to the review成員全心加入評審成員全心加入評審members

32、are fluent in architecture and architectural issues成員要訓(xùn)練有素而且口才好成員要訓(xùn)練有素而且口才好it is located close to artifacts being examined有足夠的權(quán)限檢查被評審材料有足夠的權(quán)限檢查被評審材料 26262021-12-19Preconditions: Review TeamnThe team must have access toapplicable domain knowledge (perhaps via consultants to the team)適當(dāng)?shù)念I(lǐng)域知識適當(dāng)?shù)念I(lǐng)域知識des

33、ign documents設(shè)計文檔設(shè)計文檔review criteria評審標(biāo)準(zhǔn)評審標(biāo)準(zhǔn)support staff支持人員支持人員 27272021-12-19Preconditions: OrganizationalExpectationsnA contract between the review team and the sponsor of the review must be established.Who will be told what upon completion?結(jié)束后，哪些內(nèi)容可以被告訴哪些人結(jié)束后，哪些內(nèi)容可以被告訴哪些人What will (not) be the

34、review criteria?什么是什么是(不是不是)評審標(biāo)準(zhǔn)？評審標(biāo)準(zhǔn)？What/who will be made available to team?哪些人和物是評審團(tuán)可以使用的？哪些人和物是評審團(tuán)可以使用的？What follow-up is expected?希望評審的結(jié)果如何？希望評審的結(jié)果如何？How long will the review take?要評審多久？要評審多久？ 28282021-12-19Preconditions: OrganizationalExpectationsnOrganizational culture and support must be con

35、sidered.Are reviews part of the standard project life cycle?評審是項目的基本生命周期的一部分嗎？評審是項目的基本生命周期的一部分嗎？Is serving on a review team considered good for ones career?為評審團(tuán)服務(wù)是份好差事嗎？為評審團(tuán)服務(wù)是份好差事嗎？Will the organization allow its superior designers time off from their projects to serve on review teams?企業(yè)允許它的高級設(shè)計師把他

36、的項目放在一邊來參加評審嗎？企業(yè)允許它的高級設(shè)計師把他的項目放在一邊來參加評審嗎？Is there a standing review organization (even though members may rotate on and off)?有常設(shè)的評審組織嗎有常設(shè)的評審組織嗎(即便成員常進(jìn)進(jìn)出出即便成員常進(jìn)進(jìn)出出)？ 29292021-12-19Preconditions: Read-Ahead MaterialnFor project being reviewedquestionnairesscenarios (These are created as part of the re

37、view, but time should be allowed for project members to digest the information.)nFor review teamdescription of architecturerationale for architecturenIdeally, material will address the review questions. (Passing this test because the questions were known in advance is not cheating, but good engineer

38、ing!)理想情況，材料要能說明評審時的問題理想情況，材料要能說明評審時的問題 30302021-12-19Preconditions: Ranked Qualityand Behavioral RequirementsnPlan to examine three to five quality and behavioral requirements.計劃檢查計劃檢查3到到5個質(zhì)量和行為需求個質(zhì)量和行為需求nA particular representation medium (language) for the architecture is not necessary. As long a

39、s the information can be extracted (using project members as necessary), the form of the representation is not important.一個專門的體系結(jié)構(gòu)表述媒體一個專門的體系結(jié)構(gòu)表述媒體(語言語言)目前并不是目前并不是必須的。由項目成員來講解信息，表述的形式并必須的。由項目成員來講解信息，表述的形式并不重要不重要 31312021-12-19Review ActivitiesnEvaluate“Run” the scenarios and ask the questions, or an

40、swer the items in the checklist.按場景執(zhí)行，并詢問相關(guān)問題；或者回答問卷里的問按場景執(zhí)行，并詢問相關(guān)問題；或者回答問卷里的問題題Perform the experiments.（演示實驗）（演示實驗）Execute the prototypes or simulations.運行原型或仿真程序運行原型或仿真程序nRecordcritical issues highlighted in the review評審中發(fā)現(xiàn)的嚴(yán)重問題評審中發(fā)現(xiàn)的嚴(yán)重問題comments of project members on these issues項目成員對這些問題的說明項目成員對

41、這些問題的說明 32322021-12-19Review ActivitiesnRank the issuesproject-threatening（生死攸關(guān)）（生死攸關(guān)）majorminornEach issue raised should be addressed in terms ofchanging the architecturerelaxing the requirements 33332021-12-19Review ActivitiesnWarning signsarchitecture forced to match organization組織模式影響體系結(jié)構(gòu)組織模式影響體

42、系結(jié)構(gòu)top-level components number over 25頂層組件數(shù)超過頂層組件數(shù)超過25個個one requirement drives entire design一項需求決定整個設(shè)計一項需求決定整個設(shè)計architecture depends on alternatives in the operating system體系受操作系統(tǒng)選擇的影響體系受操作系統(tǒng)選擇的影響 34342021-12-19Review ActivitiesnWarning signschoice of software components is dictated by hardware pers

43、onnel由硬件人員選擇軟件組件由硬件人員選擇軟件組件redundancy not needed for reliability面對可靠性時，不考慮冗余手段面對可靠性時，不考慮冗余手段design is exception driven零散的小的事件影響設(shè)計零散的小的事件影響設(shè)計no identifiable architect沒有確定的架構(gòu)師沒有確定的架構(gòu)師 35352021-12-19Output from ReviewnSet of ranked issuessupporting datacontained in formal reportfeedback to projectnEnha

44、nced system documentation增強的系統(tǒng)文檔增強的系統(tǒng)文檔nSet of scenarios for future use將來會用到的場景集將來會用到的場景集nIdentification of potentially reusable components認(rèn)定潛在的可重用組件認(rèn)定潛在的可重用組件nEstimation of costs and benefits估算成本和收益估算成本和收益 36362021-12-19SAAM: Software Architecture Analysis Method 一種簡單的基于場景的評審技術(shù)一種簡單的基于場景的評審技術(shù) 373720

45、21-12-19Qualities Are Too Vague for AnalysisnIs the following system modifiable?這個系統(tǒng)具有適應(yīng)性嗎？這個系統(tǒng)具有適應(yīng)性嗎？Background color of the user interface is changed merely by modifying a resource file.通過修改資源文件，可以改變界面的背景色通過修改資源文件，可以改變界面的背景色Dozens of components must be changed to accommodate a new data file format

46、.必須修改許多組件才能適應(yīng)新的文件格式必須修改許多組件才能適應(yīng)新的文件格式nA reasonable answer is一個合理的答案：一個合理的答案：yes with respect to changing background color有有，從改變背景色的角度來看，從改變背景色的角度來看no with respect to changing file format沒有沒有，從改變文件格式的角度來看，從改變文件格式的角度來看 38382021-12-19Qualities Are Too Vague for AnalysisnQualities only have meaning with

47、in a context.質(zhì)量因素只在一定的上下文里才有意義質(zhì)量因素只在一定的上下文里才有意義nSAAM specifies context through scenarios.SAAM通過場景指定上下文通過場景指定上下文 39392021-12-19ScenariosnA scenario is a brief description of a stakeholders interaction with a system.場景就是一個角色與系統(tǒng)交互過程的簡單描述場景就是一個角色與系統(tǒng)交互過程的簡單描述nWhen creating scenarios, it is important to c

48、onsider all stakeholders.建立場景的時候，一定要考慮到所有的角色建立場景的時候，一定要考慮到所有的角色Customer、User、Developer、Maintainer、AttackernThe scenarios that are of most interest in a SAAM evaluation are change scenarios.SAAM最關(guān)心的是最關(guān)心的是變化變化場景場景 40402021-12-19Steps of a SAAM EvaluationnIdentify and assemble stakeholders確認(rèn)和召集角色確認(rèn)和召集角

49、色nDevelop and prioritize scenarios制定場景，并排好優(yōu)先級制定場景，并排好優(yōu)先級nDescribe candidate architecture(s)描述候選的體系結(jié)構(gòu)描述候選的體系結(jié)構(gòu)nClassify scenarios as direct or indirect把場景分類為把場景分類為直接場景直接場景和和間接場景間接場景nPerform scenario evaluation進(jìn)行場景評估進(jìn)行場景評估nReveal scenario interactions展現(xiàn)場景間的相互影響展現(xiàn)場景間的相互影響nGenerate overall evaluation整體評

50、估整體評估 41412021-12-19Step 1: Identify and Assemble StakeholdersStakeholderInterestCustomerSchedule and budget; usefulness of system; meeting customers (or markets) expectationsEnd UserFunctionality, usabilityDeveloperClarity and completeness ofarchitecture; high cohesion andlimited coupling of parts;

51、clear interaction mechanismsMaintainerMaintainability; ability to locateplaces of change 42422021-12-19Step 1: Identify and Assemble StakeholdersStakeholderInterestSystem AdministratorEase in finding sources of operational problemsNetwork AdministratorNetwork performance, predictabilityIntergratorCl

52、arity and completeness ofarchitecture; high cohesion andlimited coupling of parts;clear interaction mechanisms 43432021-12-19Step 1: Identify and AssembleStakeholdersStakeholderInterestTesterIntegrated, consistent error handling; limited component coupling; high component cohesion; conceptual integr

53、ityApplication BuilderArchitectural clarity, completeness; interaction mechanisms; simple tailoring mechanismsRepresentative of the domainInteroperability 44442021-12-19Step 2: Stakeholders Develop andPrioritize ScenariosnScenarios should be typical of the kinds of evolution that the system must sup

54、port:場景應(yīng)該是系統(tǒng)必須支持的各種演化的典型：場景應(yīng)該是系統(tǒng)必須支持的各種演化的典型：functionalitydevelopment activitieschange activitiesnScenarios should represent tasks relevant to all stakeholders.所有角色都應(yīng)該能在場景中體現(xiàn)出來所有角色都應(yīng)該能在場景中體現(xiàn)出來nRule of thumb: 10-15 prioritized scenarios應(yīng)該有應(yīng)該有10-15個優(yōu)先的場景個優(yōu)先的場景 45452021-12-19Step 3: Describe CandidateA

55、rchitecturesnIt is frequently necessary to elicit appropriate architectural descriptions.通常，對體系結(jié)構(gòu)進(jìn)行適當(dāng)?shù)拿枋鍪欠浅Ｍǔ?，對體系結(jié)構(gòu)進(jìn)行適當(dāng)?shù)拿枋鍪欠浅Ｖ匾闹匾膎Structures chosen to describe the architecture will depend on the type of qualities to be evaluated.以我們需要的質(zhì)量為依據(jù)，選擇用來描述以我們需要的質(zhì)量為依據(jù)，選擇用來描述體系結(jié)構(gòu)的結(jié)構(gòu)體系結(jié)構(gòu)的結(jié)構(gòu) 46462021-12-19Ste

56、p 4: Classify ScenariosnThere are two classes of scenarios.Direct scenarios are those that can be executed by the system without modification.直接場景是那些不需要修改系統(tǒng)就能執(zhí)行的場景直接場景是那些不需要修改系統(tǒng)就能執(zhí)行的場景Indirect scenarios are those that require modifications to the system.間接場景是那些需要修改系統(tǒng)的場景間接場景是那些需要修改系統(tǒng)的場景nThe classific

57、ation depends upon both the scenario and the architecture.場景的類型因系統(tǒng)而異場景的類型因系統(tǒng)而異nFor indirect scenarios we gauge the order of difficulty of each change: e.g. a person-day, person-week, person-month, person-year.對間接場景，我們要度量適應(yīng)每個變化的難度（工作量）對間接場景，我們要度量適應(yīng)每個變化的難度（工作量） 47472021-12-19Step 5: Perform Scenario E

58、valuationnFor each indirect scenarioidentify the components, data connections, control connections, and interfaces that must be added, deleted, or modified明確其組件、數(shù)據(jù)連結(jié)件、控制連結(jié)件和要被增加、刪除或修改明確其組件、數(shù)據(jù)連結(jié)件、控制連結(jié)件和要被增加、刪除或修改的接口的接口estimate the difficulty of modification估算修改的難度估算修改的難度nDifficulty of modification is

59、 elicited from the architect and is based on the number of components to be modified and the effect of the modifications.修改的難度體現(xiàn)在被修改的組件個數(shù)和修改的效果修改的難度體現(xiàn)在被修改的組件個數(shù)和修改的效果nA monolithic system will score well on this step, but not on next step.只有一個模塊的系統(tǒng)在此項評估可以得到高分，但是只有一個模塊的系統(tǒng)在此項評估可以得到高分，但是 48482021-12-19St

60、ep 6: Reveal Scenario InteractionsnWhen multiple indirect scenarios affect the same components, this could indicate a problem.當(dāng)多個間接場景影響同一個組件時，可能引發(fā)問題當(dāng)多個間接場景影響同一個組件時，可能引發(fā)問題could be good, if scenarios are variants of each other可能是好事，當(dāng)場景是彼此的變體可能是好事，當(dāng)場景是彼此的變體nchange background color to greennchange backg

61、round color to redcould be bad, indicating a potentially poor separation of concerns可能是壞事，說明模塊劃分有問題可能是壞事，說明模塊劃分有問題nchange background color to rednport system to a different platform 49492021-12-19Step 7: Generate Overall EvaluationnNot all scenarios are equal.場景生來不平等場景生來不平等nThe organization must determine which scenarios are most important.必須明確哪些場景是最重要的必須明確哪些場景是最重要的nThen the organization must decide as to whether the design is acceptable “as is” or if it must be modified.評審團(tuán)得出結(jié)論，當(dāng)前的體系結(jié)構(gòu)設(shè)計可行，還評審團(tuán)得出結(jié)論，當(dāng)前的體系結(jié)構(gòu)設(shè)計可行，還是必須要修改是必須要修改