P. Clements, L. Bass, R. Kazman, G. Abowd. "Predicting Software Quality by Architecture-Level Evaluation." Proceedings of the International Conference on Software Quality, Austin, Texas, October 1995.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....kinds of interactions that may take place between them. One of the significant advantages of architectural descriptions is that they provide opportunities for analysis, including system consistency checking [3] conformance to architectural style constraints [1] conformance to quality attributes [7], and dependence analysis [30] We can model our example using a client server architectural style. The architectural style provides definitions for client, server, and server group components and the connections between them. Properties include those required for queuing theoretic performance ....

Clements, P., Bass, L., Kazman, R., Abowd, G. Predicting Software Quality by Architecture -Level Evaluation. In Proceedings of the Fifth International Conference on Software Quality, Austin, TX, October 1995.

....concerns about performance interoperability, prototyping, and reuse. 5. Analysis: Architectural descriptions provide new opportunities for analysis, including system consistency checking [2, 25] conformance to constraints imposed by an architectural style [1] conformance to quality attributes [9], dependence analysis [42] and domain specific analyses for architectures built in specific styles [10, 15, 26] 6. Management: Experience has shown that successful projects view achievement of a viable software architecture as a key milestone in an industrial software development process. ....

....and academia have proposed formal notations for representing and analyzing architectural designs. Generically referred to as Architecture Description Languages (ADLs) these notations usually provide both a conceptual framework and a concrete syntax for characterizing software architectures [9, 30]. They also typically provide tools for parsing, displaying, compiling, analyzing, or simulating architectural descriptions. Examples of ADLs include Adage [10] Aesop [15] C2 [28] Darwin [26] Rapide [25] SADL [32] UniCon [39] Meta H [6] and Wright [3] While all of these languages are ....

P. Clements, L. Bass, R. Kazman and G. Abowd. Predicting software quality by architecture-level evaluation. In Proceedings of the Fifth International Conference on Software Quality, Austin, Texas, Oct, 1995.

....2. 2 Extra functional properties In addition to functionality and structure, architectural specifications must be capable of expressing extra functional properties related to performance, capacity, environmental assumptions, and global properties such as reliability and security [MCN92, Shaw85, CBKA95] Many of these additional proper ties are qualitative, so they may require different kinds of support from more formal specifications. These other properties include: time requirements precision and accuracy timing variability reliability real time response robustness latency security ....

Paul Clements, Len Bass, Rick Kazman, and Gregory Abowd. Predicting software Qual - ity by architecture-level evaluation. In Proc Fifth International Conf on Software Quality, October 1995

.... architecture of a complex software system is a significant issue in the proper design, analysis and possibility of reuse of programs [GP95] The extraction of the architecture of programs is an essential step in applying architecture based techniques for design [GS93, LKA 95] analysis [AG94, CBKA95, LVB 93] and reuse [GAO95] The automatic extraction of architectures of Ada programs and testing for conformance to a rapide reference architecture facilitate the design, analysis and reuse of Ada programs. rapide is an event based, concurrent, object oriented language designed for ....

P. Clements, L. Bass, R. Kazman, and G. Abowd. Predicting software quality by architecture-level evaluation. In Proceedings of the 6th International Conference on Software Quality, October 1995.

.... we say that a system that can work on a different hardware platform by simply being recompiled, but requires dozens of changes to accommodate a different data type, is a modifiable system [4] For a solution to the problem of evaluation we use SAAM: the Software Architecture Analysis Method ( 8] [7]) SAAM carries out context based evaluation through the use of task scenarios. These scenarios describe specific tasks that will be carried out by various people involved with the system: users, developers, maintainers, administrators, etc. and task specified. The work described in this paper is ....

R. Kazman P. Clements, L. Bass and G. Abowd. "Predicting Software Quality by Architecture-Level Evaluation". Proceedings of the 5th International Conference on Software Quality, (Austin, TX) (to appear),

....Extra functional Properties. In addition to functionality and packaging, architectural specifications must be capable of expressing extra functional properties related to performance, capacity, environmental assumptions, and global properties such as reliability and security [Sha85, MCN92, CBKA95] Many of these additional properties are qualitative, so they may require different kinds of support from more formal specifications. These other properties include: time requirements ease of use timing variability reliability real time response robustness latency service capacity (e.g. # of ....

Paul Clements, Len Bass, Rick Kazman, and Gregory Abowd. Predicting software quality by architecture-level evaluation. In Proceedings of the Fifth International Conference on Software Quality, Austin, Texas, October 1995.

....about their properties. The promise of software architectures is that they will reduce the costs and improve the quality of software development by enabling component based construction of large scale software. Methods for the evaluation of software architectures such as SAAM [Kazman et al. 1994, Clements et al. 1995, Abowd et al. 1994] give the designer an early chance to evaluate a design based on an architectural description. The advantages of this and similar approaches are twofold. First, they are applicable at the design level as opposed to the source level. Second, they can be used early in the design ....

....anticipated usage. Thus, when we speak about a quality such as modifiability, we can only discuss designing for a specific set of anticipated modifications. Kazman et al. have developed an evaluation method called SAAM (System Architecture Analysis Method) Kazman et al. 1994, Abowd et al. 1994, Clements et al. 1995] that by using concrete benchmarks against which a software architecture can be tested forces the designer to clarify the meaning of terms like modifiability, reliability, performance, etc. SAAM is based on the notion of scenarios. A scenario is a description of the anticipated usage of ....

Paul Clements, Len Bass, Rick Kazman, and Gregory Abowd "Predicting Software Quality by Architecture-Level Evaluation", Proceedings, Fifth International Conference on Software Quality. Austin, Tx, October 1995

....these domain scenarios and candidate architectures, and a score is assigned to the degree by which a candidate architecture satisfies the expectations of each scenario. Candidate architectures can then be contrasted in terms of their fulfillment of scenario based expectations of them [Abowd 94, Clements 95a, Kazman 95] 3.2.5 Architecture Can Be the Basis for Training The structure, plus a high level description of how the components interact with each other to carry out the required behavior, often serves as the high level introduction to the system for new project members. 3.2.6 An Architecture ....

....described, as well as their primary intended audiences and issues. It was adapted from a similar diagram shown in Kruchten [Kruchten 95] Figure 4: Architectural Views Scenarios Scenarios are scripts of individual pieces of a system s functionality. Scenarios are useful in analyzing a given view [Clements 95a] or in showing that the elements of multiple views work together properly [Kruchten 95] We can think about the scenar End Users . functionality Programmers . software management System integrators . performance . scalability . throughput System engineers . system topology . delivery . ....

Clements, P.; Bass, L.; Kazman, F; & Abowd, G. "Predicting Software Quality by Architecture-Level Evaluation," 485498. Proceedings, Fifth International Conference on Software Quality. Austin, TX, Oct. 23-26, 1995. Austin, TX: American Society for Quality Control, Software Division, 1995.

....and the interactions (called connections) between those elements. Software architecture has emerged as an important sub field within software engineering in the past few years ( 13] 19] and our particular concern has been in the use of scenarios to analyze architectural descriptions ([9], 15] High level system descriptions are useful for communicating to a large audience the general behavior of a system, so it is important that these architectural descriptions be understandable and informative. In addition, designers make claims about qualities that their systems possess (e.g. ....

P. Clements, L. Bass, R. Kazman, G. Abowd, "Predicting Software Quality by Architecture-Level Evaluation", In the proceedings of the International Conference on Software Quality, Austin, TX, October 1995. To appear.

....to the same problem. We performed a SAAM evaluation on KWIC to determine if we could reproduce similar results as have been previously published. We actually found that our own analysis provided much more rationale for deciding between solutions to this problem than was previously published [2]. 4. Embedded audio system An automotive company which develops their own embedded audio systems used a scenario based technique during the design of the next generation of audio systems. 5. Visual debuggers A SAAM evaluation comparing two public domain visual debuggers. For each of the ....

....led to the customer s decision to not purchase the supplier s system, and this decision was based on concrete evidence that the system did not directly support a large enough number of required scenarios. It is very important to point out that this C1 C4 C5 D3 C7 C9 C10 [1] C: init ops [2] D: bxr parms [3] C: invoke bxr [4] D: get ids [5] D: do cty [6] C: invoke ccj Figure 4: Sample flow description of direct scenarios [7] D: ccj parms C1 D3 Passive data Computation Components Active data C7 D1 C5 C8 C3 C4 C2 C6 C9 C10 Figure 5: Static architecture with connections D2 ....

Clements, P., Bass, L., Kazman, R., Abowd, G., "Predicting Software Quality by ArchitectureLevel Evaluation", 5th International Conference on Software Quality, (Austin, TX), October 1995, to appear.

....on an existing system s architecture. Our interest here is in the analysis of architectural descriptions in order to determine how well they satisfy a large class of quality criteria. We are extending our scenario based evaluation technique, the Software Architecture Analysis Method (SAAM) [3][7] 8] to perform this analysis. A reverse engineering process for extracting architectural information from an existing system. Beyond the traditional structural analyses provided by existing commercial and research tools, MORALE will extract behavior information derived from dynamic analysis ....

Clements, P., Bass, L., Kazman, R., and Abowd, G. Predicting software quality by architecture-level evaluation. In Proceedings of the Fifth International Conference on Software Quality, Austin, Texas, October 1995.

....case study appears in the next section. We have also conducted a number of other industrial and academic case studies in scenario based analysis as SAAM was maturing in areas such as user interface development environments [11] internet information systems [13] key word in context (KWIC) systems [3], embedded audio systems, and visual debuggers. 3.1 The WRCS System 3.1.1 System context purpose In this section we will discuss the application of SAAM to a commercially available revision control system, based upon RCS [18] which we will call WRCS. WRCS provides the functionality to allow ....

Clements, P., Bass, L., Kazman, R., Abowd, G., "Predicting Software Quality by ArchitectureLevel Evaluation", 5th International Conference on Software Quality, (Austin, TX), October 1995, to appear.

No context found.

P. Clements, L. Bass, R. Kazman, G. Abowd. "Predicting Software Quality by Architecture-Level Evaluation." Proceedings of the International Conference on Software Quality, Austin, Texas, October 1995.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC