| D. Garlan, A.J. Kompanek, S.-W. Cheng. Reconciling the Needs of Architectural Description with Object Modeling Notations. Science of Computer Programming vol. 44, 2001. |
....proposed by Medvidovic et al. 25] Their proposal emphasizes the definition of suitable stereotypes to represent the peculiar concepts, but does not consider the problem of analyzing and validating designed models. Even more attention to the usability of concepts is payed by Garlan et al. in [15], where they propose several di#erent alternatives for modeling architectural structures using UML. They present a systematic approach to the use of UML and describe the strength and weaknesses of adopting a particular modeling strategy. The main goal of our work is not on rendering architectural ....
D. Garlan, S.-W. Cheng, and A. J. Kompanek. Reconciling The Needs of Architectural Description with Object-Modeling Notations. Science of Computer Programming, 44(1):23--49, July 2002.
....render software architectures. Their proposal emphasizes the definition of suitable stereotypes to represent the peculiar concepts, but does not consider the problem of analyzing and validating designed models. Even more attention to the representation of concepts is payed by Garlan et al. in [12], where they propose several di#erent alternatives for modeling architectural structures using UML. They present a systematic approach to the use of UML and describe the strength and weaknesses of adopting a particular modeling strategy. The main goal of our work is not on rendering architectural ....
D. Garlan, S.-W. Cheng, and A. J. Kompanek. Reconciling The Needs of Architectural Description with Object-Modeling Notations. Science of Computer Programming, 44(1):23--49, July 2002.
....that render them are too di#cult for the user. This is why we decided for a well known representation of concepts, paired with a formal definition of their interaction and composition. Given the UML like representation, we must mention the works by Medvidovic et al. 21] and by Garlan et al. [11]. They both study the suitability of UML to represent software architectures and identify some di#erent alternatives. They can be seen as complement instead of alternatives to our proposal: We do exploit UML class diagrams to define the style of the architecture, but we do not address the ....
D. Garlan, S.-W. Cheng, and A. J. Kompanek. Reconciling The Needs of Architectural Description with Object-Modeling Notations. Science of Computer Programming, 44(1):23--49, July 2002.
....view. The Real Time profile uses collaboration diagrams while Rumpe et al. use modified UML class diagrams arguing that the UML collaboration diagrams are intended to describe one possible interaction while the class diagrams deal with all possible paths of interaction. 2.3. 3 UML and ADLs In [21] Garlan et al. argue that UML is not suitable as is for modeling the structure of a system typically used in the architecture description languages. The key issue is a semantic mismatch between ADL abstractions and modeling concepts available in UML. However, several papers try to map various ....
D. Garlan, Shang-Wen Cheng, A. Kompanek, "Reconciling the needs of architectural description with object-modeling notations," Science of Computer Programming, Vol 44, Elsevier Science, 2002, pp 23-49.
....and reuse, as well as useful building blocks of software architectures. While the former view is supported by practical component systems [15, 19, 24] the latter view appears to be lagging behind. The current trend of modeling software architectures using UML is criticized as being inadequate [8], and while the research in component systems based on architecture description languages (ADL component systems) cites inarguable benefits of such systems [1, 5, 13, 14, 23] the described projects rarely get past research prototypes. The discrepancy between the cited benefits of ADL component ....
Garlan D., Kompanek A.: Reconciling the Needs of Architectural Description with Object-Modeling Notations, Proceedings of UML 2000, United Kingdom, 2000
....typically found in architectural description languages. Basically, the problems are the semantic clashes between the UML and ADL concepts (e.g. the term component has slightly different meanings in both worlds) the lack of a clear mapping between the UML and the ADL concepts (as discussed in [5]) the vague semantics of UML versus the precise semantics required by ADLs [10] and the gaps that currently UML has for modeling and exploiting architectural styles, explicit software connectors, and local and global architectural constraints [13] Several authors have studied the ....
.... has for modeling and exploiting architectural styles, explicit software connectors, and local and global architectural constraints [13] Several authors have studied the expressiveness of UML in order to represent software architectures, and the ways to adapt this notation to do so (in particular [5, 13]. The difficulties found made the OMG include an explicit requirement about supporting component based development and run time architectures of applications in the UML 2.0 Superstructure Request for Proposals (OMG document ad 00 09 02. However, after looking at the initial proposals received, ....
[Article contains additional citation context not shown here]
David Garlan, Shang-Wen Chen, and Andrew J. Kompanek. Reconciling the needs of architectural descriptions with object-modeling notations. To appear in Science of Computer Programming, 2002.
....implementing the architecture description. Methodologies used for the architecture description based on particular ADLs representations have been supplemented lately with a UML approach. Instead of using specific ADLs boxes and lines, UML extensions are defined for capturing the ADLs semantics [8]. An example is the C2 ADL mapping to UML [19] This trend of UML extension and adaptation for different purposes is encouraged by OMG and some Profiles are already made available [26] 3.3 Summary We are defining a UML based ADL for the design phase of a agent architecture system. RM ODP ....
....Each platform element can be represented at the type level and the instance level. An instance of a type is defined in ODP as a X (anything) that satisfies the type. A type is a predicate that characterizes collections of X s. The same approach of type and instance is referred in [8]. For simplicity, an analogy with object modeling concepts would be classes for types and instances of these classes. We model the platform elements in the MASIF DESIGN profile with stereotypes illustrated in Table 2. Table 2. The stereotypes modeling the MASIF platform elements MASIF ....
Garlan D. and Kompanek A.J., Reconciling the Needs of Architectural Description with Object-Modeling Notations, UML2000.
....in dramatically different ways. FInally, it is a relatively abstract language and thus it is not easy to communicate UML designs to non technical people. Despite these problems, it is receiving a great deal of attention from practicing architects. For a critical evaluation of UML as an ADL, see [16]. 4. Architecture as Early Design Decisions A software architecture is a coherent, justified collection of a system s earliest set of design decisions. These decisions, such as what style the system is composed of (client server, pipeand filter, distributed objects, repository centric, etc. ....
Garlan, D., Kompanek, A., "Reconciling the Needs of Architectural Description with Object-Modeling Notations", Proceedings of the Third International Conference on the Unified Modeling Language, October, 2000.
No context found.
D. Garlan, A.J. Kompanek, S.-W. Cheng. Reconciling the Needs of Architectural Description with Object Modeling Notations. Science of Computer Programming vol. 44, 2001.
No context found.
D. Garlan, A.J. Kompanek, S.-W. Cheng. Reconciling the Needs of Architectural Description with Object Modeling Notations. Science of Computer Programming vol. 44, 2001.
No context found.
D. Garlan, A.J. Kompanek, S.-W. Cheng. Reconciling the Needs of Architectural Description with Object Modeling Notations. Science of Computer Programming vol. 44, 2001.
.... There has been considerable interest in using general purpose object design notations for architectural modeling [8, 24] Moreover, recently there have been a number of proposals that attempt to show how the concepts found in ADLs can be mapped directly into an object oriented notation like UML [29, 23, 17]. These alternatives are illustrated in Figure 2. Path A D is one in which an ADL is used as the modeling language. Path B E is one in which UML is used as the modeling notation. Path A C E, is one in which an architecture is first represented in an ADL, but then transformed into UML before ....
D. Garlan, A. J. Kompanek and P. Pinto. Reconciling the needs of architectural description with objectmodeling notations. Technical report, Carnegie Mellon University, December 1999.
No context found.
D. Garlan and A. J. Kompanek, "Reconciling the Needs of Architectural Description with Object-Modeling Notations", <<UML>> 2000.
No context found.
D. Garlan and A. J. Kompanek, "Reconciling the Needs of Architectural Description with Object-Modeling Notations ", <<UML>> 2000.
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