Meiru Che

Practical experiences of cost/schedule measure through earned value management and statistical process control

Cost and schedule measures are the most important support activities for the success of a project; it provides the basis for process improvement and project management. This paper reports practical experiences on using EVM (Earned Value Management) and SPC(Statistical Process Control) in cost/ schedule measure. The analysis of experience data indicates the distributions of CPI(Cost Performance Index) and SPI(Schedule Performance Index) index are generally following the normal distribution. And consequently, it is reasonable and effective to employ SPC in EVM.

An empirical study on establishing quantitative management model for testing process

Frequently, effort of defect detecting and fixing are counted into software testing activities/phase. Current leading software estimation methods, such as COCOMO II, mainly estimate the effort depending on the size of software product and allocate testing effort proportionally. It can not predict detecting and fixing effort accurately. In fact, testing effort is significantly influenced by the quality of other software development activities. These lead to the difficulty of the testing effort to be estimated accurately. It is a challenging issue for quantitative software process management. In this paper, we propose an empirical method to identify performance objectives, establish performance baseline and establish quantitative management model for testing process. The method has been successfully applied to a software organization for their quantitative management of testing process.

Scenario-Based Architectural Design Decisions Documentation and Evolution

Software architecture is considered as a set of architectural design decisions. Capturing and representing architectural design decisions during the architecting process is necessary for reducing architectural knowledge evaporation. Moreover, managing the evolution of architectural design decisions helps to maintain consistency between requirements and the deployed system. In this paper, we create the Triple View Model (TVM) as a general architecture framework for documenting architectural design decisions. The TVM clarifies the notion of architectural design decisions in three different views and covers key features of the architecting process. Based on the TVM, we propose a scenario-based methodology (SceMethod) to manage the documentation and evolution of architectural design decisions. We also conduct a case study on an industrial project to validate the applicability and the effectiveness of the TVM and the SceMethod. The results show they provide complete documentation on architectural design decisions for creating a system architecture, and well support architecture evolution with changing requirements.

Architectural Design Decisions in Open Software Development: A Transition to Software Ecosystems

Managing architectural design decisions (ADDs) in software development process is an essential task for architectural knowledge management. As software ecosystems become a new software development paradigm in software engineering processes, it is important and necessary to capture and represent ADDs in open software development, and to evolve architectural knowledge with minimum knowledge evaporation in the open ecosystem community. So far, little work has been done on managing architectural decisions in software ecosystems in current software architecture research and practice. In this research position paper, we discuss the typical characteristics of software ecosystems which may influence architecture decision-making processes in software development, and identify the essential aspects that should be considered for managing ADDs in the context of software ecosystem. In addition, we discuss major challenges of managing ADDs for software ecosystems, and we also propose possible directions in research to solve the problems.

An approach to documenting and evolving architectural design decisions

Software architecture is considered as a set of architectural design decisions (ADDs). Capturing and representing ADDs during the architecting process is necessary for reducing architectural knowledge evaporation. Moreover, managing the evolution of ADDs helps to maintain consistency between requirements and the deployed system. In this work, we create the Triple View Model (TVM) as a general architecture framework for documenting ADDs. The TVM clarifies the notion of ADDs in three different views and covers key features of the architecting process. Based on the TVM, we propose a scenario-based method (SceMethod) to manage the documentation and the evolution of ADDs. Furthermore, we also develop a UML metamodel that incorporates evolution-centered characteristics to manage evolutionary architectural knowledge. We conduct a case study to validate the applicability and the effectiveness of our model and method. In our future work, we plan to investigate how to support ADD documentation and evolution in geographically separated software development (GSD).

Regression test selection for Android applications

Mobile platform pervades human life, and much research in recent years has focused on improving the reliability of mobile applications on this platform, for example by applying automatic testing. However, researchers have primarily considered testing of single version of mobile applications. Although regression testing has been extensively studied for desktop applications, the approaches for desktop applications cannot be directly applied to mobile applications. Our approach leverages the combination of static impact analysis and dynamic code coverage information, and identifies a subset of test cases for re-execution on the modified app version. We implement our approach for Android apps, and illustrate its usefulness based on an Android application.

Redroid: A Regression Test Selection Approach for Android Applications

As the mobile platform pervades human life, much research in recent years has focused on improving the reliability of mobile applications on this platform, for example by applying automatic testing. However, researchers have primarily consid- ered testing of single version of mobile applications. Although regression testing has been extensively studied for desktop appli- cations, and many efficient and effective approaches have been proposed, these approaches cannot be directly applied to mobile applications. We first present a bug study on real-world Android bugs to show the existence of regression bugs, which motivates the need for an efficient regression test selection technique for Android applications. Next, we introduce Redroid, a new ap- proach to regression test selection for Android applications. Our approach leverages the combination of static impact analysis and dynamic code coverage, and identifies a subset of test cases for re- execution on the modified application version. We implement our approach for Android applications, and demonstrate its efficacy through an extensive empirical study. as a runtime environment, which is significantly different from the Java Virtual Machine. This paper first presents a bug study based on 10 real- world Android apps from Google Code Repository (3). The study shows that there are regressions for Android apps during evolution, and an efficient and effective regression testing approach is highly needed for this area. This paper then intro- duces Redroid, a new approach to regression test selection for Android apps. Given a test suite that was performed on the original Android app, and the two versions involved in a change, Redroid identifies a subset of the tests that must be re-executed to test the new Android version. Leveraging the combination of static impact analysis with coverage informa- tion that is dynamically generated at runtime, our approach identifies a subset of tests to check the behaviors of the modified version that can potentially be different from the original version. We developed a prototype tool for Redroid, and conducted an evaluation based on two real-world Android apps, which shows that our approach can significantly reduce the number of tests for re-execution after an Android app is modified.

An Evaluation Study of Architectural Design Decision Paradigms in Global Software Development.

Global software development (GSD) is considered as the coordinated activities of software development that are geographically and temporally distributed. The management of architectural knowledge, specifically, architectural design decisions (ADDs), becomes important in GSD due to the geographical, temporal, and cultural challenges in global envi- ronments. Based on our previous work on ADD management in localized software development (LSD), we present five ADD paradigms used for GSD projects with different organizational structures. We also investigate the benefits and the challenges of the ADD paradigms by conducting an evaluation of the paradigms using extensive archived semi-structured interview data from industrial GSD projects. We aim to provide a fundamental framework for managing ADD documentation and evolution in GSD, as well as offer useful insights into managing architectural knowledge in a global setting. Keywords-architectural design decisions; global software de- velopment; documentation; evolution I. INTRODUCTION Global software development (GSD) is an increasing fo- cus in the field of software engineering. It can be considered as the coordinated activities of software development that are not localized and centralized but geographically and temporally distributed (12). Little attention has been paid to software architecting processes and software architectural knowledge management in the context of GSD. Similar to localized software projects, software architecting and architectural knowledge are important to support design- ing, developing, testing, and evolving software. We note, however, that in the global development of large complex systems, architecture plays an even more critical role in the structure of the project (11). Therefore, managing and coordinating architectural knowledge such as architectural design decisions (ADDs) is a significant and also relatively new research problem in the context of GSD. In our previous work on ADD management, we had an overall goal of providing a systematic approach that supports ADD documentation and evolution in a localized software development (LSD) context. Based on this, in this paper, we present and discuss five typical ADD management paradigms for global software projects, and we also conduct an evaluation on these paradigms using archived semi- structured interview data from industrial GSD projects to investigate the benefits and the challenges of each paradigm in the GSD contexts. Since little work has been done on ADD documentation and evolution in GSD research and practice, we aim to provide a fundamental framework for managing ADD documentation and evolution in a global setting, and also provide better insights into architectural knowledge management for researchers and practitioners in GSD contexts in the field of software architecture. To the best of our knowledge, our study is the first to provide ADD management paradigms in GSD projects and to support architectural knowledge management in global settings. Our study provides evidence that managing ADDs in the GSD contexts reduces the complexity of coordination and integration among multiple distributed sites, decreases misunderstanding among different people, and also offers useful documentation for project planning and other man- agement policies. Our evaluation is also the first industrial investigation into the benefits and the challenges of global ADD management in practice.

Evaluating Architectural Design Decision Paradigms in Global Software Development

Global software development (GSD) is considered as the coordinated activities of software development that are geographically and temporally distributed. The management of architectural knowledge, specifically, architectural design decisions (ADDs), becomes important in GSD due to the geographical, temporal, and cultural challenges in global environments. Based on our previous work on ADD management in localized software development, we present five ADD paradigms for GSD projects with different organizational structures. We also investigate the benefits and the challenges of these ADD paradigms by conducting an evaluation of the paradigms using extensive archived semi-structured interview data from industrial GSD projects. We aim to provide a fundamental framework for managing ADD documentation and evolution in GSD, as well as offer useful insights into managing architectural knowledge in a global setting.