Software architecture is a recognized and indispensable part of system development. Software architecture is often defined in terms of components and connectors, or the "high-level conception of a system". In recent years, there has been an awareness that not only the design itself is important to capture, but also the knowledge that has led to this design. This so-called architectural knowledge concerns the set of design decisions and their rationale. Capturing architectural knowledge is difficult. Part of it is tacit and difficult to verbalize. Like developers, software architects are not inclined to document their solutions. Establishing ways to effectively manage and organize architectural knowledge is one of the key challenges of the field of software architecture. This architectural knowledge plays a role during development, when architects, developers, and other stakeholders must communicate about the system to be developed, possibly in a global setting. It also plays a role during the evolution of a system, when changes are constrained by decisions made earlier.
{"title":"Software Architecture Knowledge Management","authors":"H. Vliet","doi":"10.1109/ASWEC.2008.69","DOIUrl":"https://doi.org/10.1109/ASWEC.2008.69","url":null,"abstract":"Software architecture is a recognized and indispensable part of system development. Software architecture is often defined in terms of components and connectors, or the \"high-level conception of a system\". In recent years, there has been an awareness that not only the design itself is important to capture, but also the knowledge that has led to this design. This so-called architectural knowledge concerns the set of design decisions and their rationale. Capturing architectural knowledge is difficult. Part of it is tacit and difficult to verbalize. Like developers, software architects are not inclined to document their solutions. Establishing ways to effectively manage and organize architectural knowledge is one of the key challenges of the field of software architecture. This architectural knowledge plays a role during development, when architects, developers, and other stakeholders must communicate about the system to be developed, possibly in a global setting. It also plays a role during the evolution of a system, when changes are constrained by decisions made earlier.","PeriodicalId":231903,"journal":{"name":"19th Australian Conference on Software Engineering (aswec 2008)","volume":"86 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124398618","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper seeks to combine a traditional computer science course with software engineering process. A computer science course such as computer graphics, digital image processing, compiler or operating systems gives a chance to students working in a group to write a realistic large project of almost 3000 lines and make it function correctly. The course involves a team programming project. The software engineering lifecycle addresses issues on design, coding, documenting, unit testing and integration testing. Innovative and active learning based educational methods are adopted for teaching such a course.
{"title":"Software Engineering Practice in Computer Science Courses","authors":"S. Chatterjee","doi":"10.1109/ASWEC.2008.58","DOIUrl":"https://doi.org/10.1109/ASWEC.2008.58","url":null,"abstract":"This paper seeks to combine a traditional computer science course with software engineering process. A computer science course such as computer graphics, digital image processing, compiler or operating systems gives a chance to students working in a group to write a realistic large project of almost 3000 lines and make it function correctly. The course involves a team programming project. The software engineering lifecycle addresses issues on design, coding, documenting, unit testing and integration testing. Innovative and active learning based educational methods are adopted for teaching such a course.","PeriodicalId":231903,"journal":{"name":"19th Australian Conference on Software Engineering (aswec 2008)","volume":"57 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129446810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Summary form only given. No change happens in isolation, and the rate of change can have a significant impact on the acceptance level of a change. This paper reviews the interrelationships between technology, people at the micro, and macro level of organisational structure, business goals, and the external business environment as they interact within an IT change cycle. Mixing personal experience with industry case studies this paper examines how the implementation of a change cycle has an immediate feedback into the continued development of the change cycle; it demonstrates that human factors and communication mechanisms have more of an impact on the success level of a change programme than the selection of the 'right' technology or being an early adopter of new development techniques. Focusing on change within the software and IT domain examples are reviewed from the arenas of IT gaming, air traffic control, and general large-scale complex system integration change cycles. Using the Change Cycle definition developed by Brock and Salerno as a reference structure this paper examines each of the six key phases of the change cycle as they relate to staff and adoption of technical evolutions.
{"title":"Targeted Change. Are You Aready Irrelevant?","authors":"M. Noble","doi":"10.1109/ASWEC.2008.1","DOIUrl":"https://doi.org/10.1109/ASWEC.2008.1","url":null,"abstract":"Summary form only given. No change happens in isolation, and the rate of change can have a significant impact on the acceptance level of a change. This paper reviews the interrelationships between technology, people at the micro, and macro level of organisational structure, business goals, and the external business environment as they interact within an IT change cycle. Mixing personal experience with industry case studies this paper examines how the implementation of a change cycle has an immediate feedback into the continued development of the change cycle; it demonstrates that human factors and communication mechanisms have more of an impact on the success level of a change programme than the selection of the 'right' technology or being an early adopter of new development techniques. Focusing on change within the software and IT domain examples are reviewed from the arenas of IT gaming, air traffic control, and general large-scale complex system integration change cycles. Using the Change Cycle definition developed by Brock and Salerno as a reference structure this paper examines each of the six key phases of the change cycle as they relate to staff and adoption of technical evolutions.","PeriodicalId":231903,"journal":{"name":"19th Australian Conference on Software Engineering (aswec 2008)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129655894","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Design patterns are programming language and project independent design artefacts that are widely used by professional software engineers. They represent best practise software design. However, the fact that design patterns are highly abstract makes it difficult for undergraduate students to understand and apply them. In this paper, we present DPLab, a tool we have developed to assist students in recognising and comprehending design patterns.
{"title":"Tool Support for Teaching Design Patterns","authors":"Jens Dietrich, E. Kemp","doi":"10.1109/ASWEC.2008.47","DOIUrl":"https://doi.org/10.1109/ASWEC.2008.47","url":null,"abstract":"Design patterns are programming language and project independent design artefacts that are widely used by professional software engineers. They represent best practise software design. However, the fact that design patterns are highly abstract makes it difficult for undergraduate students to understand and apply them. In this paper, we present DPLab, a tool we have developed to assist students in recognising and comprehending design patterns.","PeriodicalId":231903,"journal":{"name":"19th Australian Conference on Software Engineering (aswec 2008)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130494540","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We analyzed the major differences between closed and open source software development from a software reliability perspective. We examined real world bug report data from six open source software projects, as well as the relationship between the release cycles and the stability characteristics of the bug report data of open source software projects. We then modeled the bug report data using nonparametric techniques. The experimental results suggest that generalized additive models and exponential smoothing approaches are suitable for the estimation of software reliability at least for some of the open source projects.
{"title":"Analyzing and Modeling Open Source Software Bug Report Data","authors":"Fengzhong Zou, Joseph G. Davis","doi":"10.1109/ASWEC.2008.13","DOIUrl":"https://doi.org/10.1109/ASWEC.2008.13","url":null,"abstract":"We analyzed the major differences between closed and open source software development from a software reliability perspective. We examined real world bug report data from six open source software projects, as well as the relationship between the release cycles and the stability characteristics of the bug report data of open source software projects. We then modeled the bug report data using nonparametric techniques. The experimental results suggest that generalized additive models and exponential smoothing approaches are suitable for the estimation of software reliability at least for some of the open source projects.","PeriodicalId":231903,"journal":{"name":"19th Australian Conference on Software Engineering (aswec 2008)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134352950","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Model transformations form a key part of MDA (model-driven architecture). Most of the studies deal with the transformations from PIM (platform-independent model) to PSM (platform-specific model) and PSM to Code, but very few deal with the transformation from CIM (computation-independent model) to PIM. This last transformation usually depends on business analysts' and software architects' experience and creativity. This paper proposes a disciplined approach to transform a CIM into a PIM. It first uses UML2 activity diagrams to model the business processes up to the users' tasks. The activity diagrams are then detailed to specify the system requirements. The system components are directly deduced from the requirement model elements. Finally, a set of business archetypes helps detail the system components to yield the PIM. The same approach applies equally to CIM and PIM built to model inter-enterprise processes and systems. A case study illustrates our approach. It demonstrates how it reinforces the components traceability and reusability and how it globally improves the modeler's efficiency. Furthermore, the use of the activity diagrams, as a single technique to build business process and requirement models, is an important facilitator which prepares our further work to automate this approach.
{"title":"Transformation from CIM to PIM Using Patterns and Archetypes","authors":"S. Kherraf, E. Lefebvre, W. Suryn","doi":"10.1109/ASWEC.2008.63","DOIUrl":"https://doi.org/10.1109/ASWEC.2008.63","url":null,"abstract":"Model transformations form a key part of MDA (model-driven architecture). Most of the studies deal with the transformations from PIM (platform-independent model) to PSM (platform-specific model) and PSM to Code, but very few deal with the transformation from CIM (computation-independent model) to PIM. This last transformation usually depends on business analysts' and software architects' experience and creativity. This paper proposes a disciplined approach to transform a CIM into a PIM. It first uses UML2 activity diagrams to model the business processes up to the users' tasks. The activity diagrams are then detailed to specify the system requirements. The system components are directly deduced from the requirement model elements. Finally, a set of business archetypes helps detail the system components to yield the PIM. The same approach applies equally to CIM and PIM built to model inter-enterprise processes and systems. A case study illustrates our approach. It demonstrates how it reinforces the components traceability and reusability and how it globally improves the modeler's efficiency. Furthermore, the use of the activity diagrams, as a single technique to build business process and requirement models, is an important facilitator which prepares our further work to automate this approach.","PeriodicalId":231903,"journal":{"name":"19th Australian Conference on Software Engineering (aswec 2008)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133583445","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents an experience gained in the development of a large component based application that handles massive federal and state government labor market information data. The process of transforming user requirements to deliverable software components are shown in various stages of development. Components developed are rated against existing specifications. Further, these components are classified based on their functionalities and location. Statistical information is presented for components, component types and re-use types. Our experience suggests two practical componentization methodologies and JavaBeans extensions. As per industry-taste, these extensions can become a part of future specifications in order to enable the development of fully component based applications. We also demonstrate uses of patterns to document our architecture.
{"title":"Component Based Software Development - An Industrial Experience with a Labor Market Information System","authors":"B. Upadhyaya","doi":"10.1109/ASWEC.2008.17","DOIUrl":"https://doi.org/10.1109/ASWEC.2008.17","url":null,"abstract":"This paper presents an experience gained in the development of a large component based application that handles massive federal and state government labor market information data. The process of transforming user requirements to deliverable software components are shown in various stages of development. Components developed are rated against existing specifications. Further, these components are classified based on their functionalities and location. Statistical information is presented for components, component types and re-use types. Our experience suggests two practical componentization methodologies and JavaBeans extensions. As per industry-taste, these extensions can become a part of future specifications in order to enable the development of fully component based applications. We also demonstrate uses of patterns to document our architecture.","PeriodicalId":231903,"journal":{"name":"19th Australian Conference on Software Engineering (aswec 2008)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117163485","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Following changes in business processes and technology enterprises often need to interoperate protocols of heterogeneous middleware. This paper presents an automated, data-driven approach to protocol-level middleware interoperation, based on declarative descriptions of these protocols. The run-time engine of TUBE (the ubiquitous broker environment), provides a layer between application-level components and middleware that allows a degree of independence from particular middleware implementations by taking protocol descriptions and performing the necessary translations. The TUBE framework is designed to try alternatives if the searched protocol is unavailable, making it easy to deploy different types of middleware and special protocol extensions. TUBE has been implemented and successfully tested across a range of commonly used middleware, including synchronous, asynchronous, object-based, binary and text-based protocols. Key components of the system are currently in operation in a large Australian corporation.
{"title":"TUBE: Automated Protocol-Level Middleware Interoperation","authors":"R. Slamkovic, G. Fernandez, J. McGovern","doi":"10.1109/ASWEC.2008.19","DOIUrl":"https://doi.org/10.1109/ASWEC.2008.19","url":null,"abstract":"Following changes in business processes and technology enterprises often need to interoperate protocols of heterogeneous middleware. This paper presents an automated, data-driven approach to protocol-level middleware interoperation, based on declarative descriptions of these protocols. The run-time engine of TUBE (the ubiquitous broker environment), provides a layer between application-level components and middleware that allows a degree of independence from particular middleware implementations by taking protocol descriptions and performing the necessary translations. The TUBE framework is designed to try alternatives if the searched protocol is unavailable, making it easy to deploy different types of middleware and special protocol extensions. TUBE has been implemented and successfully tested across a range of commonly used middleware, including synchronous, asynchronous, object-based, binary and text-based protocols. Key components of the system are currently in operation in a large Australian corporation.","PeriodicalId":231903,"journal":{"name":"19th Australian Conference on Software Engineering (aswec 2008)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123515943","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Microkernel-based approach provides operating system support for developing embedded systems with high performance and safety through memory protection. This allows us to introduce architectural mechanisms that enable good separation of concerns, while still satisfy the performance, security and reliability needs of embedded systems. Organizing embedded software as interacting components with well-defined interfaces is compatible with modeling methods. The challenge issue is to seamlessly transform models and integrate tools at different levels of design, implementation and deployment. In this paper we present our solution to this issue. Our contribution is twofold: first, we device a unified model driven and component based development approach. We adopt a layered architecture to construct a tool chain, which allows flexible extension at different layers. Second, we develop a software tool suite in order to support and demonstrate our solution, which includes a UML-based modeling environment and a set of component development tools on top of a microkernel operating system.
{"title":"Component Architecture and Modeling for Microkernel-Based Embedded System Development","authors":"Y. Liu, T. Wong","doi":"10.1109/ASWEC.2008.11","DOIUrl":"https://doi.org/10.1109/ASWEC.2008.11","url":null,"abstract":"Microkernel-based approach provides operating system support for developing embedded systems with high performance and safety through memory protection. This allows us to introduce architectural mechanisms that enable good separation of concerns, while still satisfy the performance, security and reliability needs of embedded systems. Organizing embedded software as interacting components with well-defined interfaces is compatible with modeling methods. The challenge issue is to seamlessly transform models and integrate tools at different levels of design, implementation and deployment. In this paper we present our solution to this issue. Our contribution is twofold: first, we device a unified model driven and component based development approach. We adopt a layered architecture to construct a tool chain, which allows flexible extension at different layers. Second, we develop a software tool suite in order to support and demonstrate our solution, which includes a UML-based modeling environment and a set of component development tools on top of a microkernel operating system.","PeriodicalId":231903,"journal":{"name":"19th Australian Conference on Software Engineering (aswec 2008)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116814543","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Distributed transaction processing is the best solution for providing enterprise software systems with high performance. However, it needs an environment that assures consistency beside high concurrency. Therefore in this paper we extend the concurrency control service (CCS) of CORBA with new features, such as generic calling context, fault tolerance and deadlock management. The most outstanding feature of ECCS is the generic calling context that makes it applicable for both transactional and non-transactional clients. Through fault tolerance considerations, ECCS supports long-duration distributed transactions in addition to short-duration ones. Deadlock management efficiently prevents any deadlock occurrences. The experimentations show that we successfully included all the new capabilities with an acceptable amount of overhead and without negatively affecting the scalability of CCS.
{"title":"An Extended Concurrency Control Service for CORBA","authors":"Nima Sharifimehr, S. Sadaoui","doi":"10.1109/ASWEC.2008.14","DOIUrl":"https://doi.org/10.1109/ASWEC.2008.14","url":null,"abstract":"Distributed transaction processing is the best solution for providing enterprise software systems with high performance. However, it needs an environment that assures consistency beside high concurrency. Therefore in this paper we extend the concurrency control service (CCS) of CORBA with new features, such as generic calling context, fault tolerance and deadlock management. The most outstanding feature of ECCS is the generic calling context that makes it applicable for both transactional and non-transactional clients. Through fault tolerance considerations, ECCS supports long-duration distributed transactions in addition to short-duration ones. Deadlock management efficiently prevents any deadlock occurrences. The experimentations show that we successfully included all the new capabilities with an acceptable amount of overhead and without negatively affecting the scalability of CCS.","PeriodicalId":231903,"journal":{"name":"19th Australian Conference on Software Engineering (aswec 2008)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123967270","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：