Hong Mei

Towards automatic model synchronization from model transformations

The metamodel techniques and model transformation techniques provide a standard way to represent and transform data, especially the software artifacts in software development. However, after a transformation is applied, the source model and the target model usually co-exist and evolve independently. How to propagate modifications across models in different formats still remains as an open problem. In this paper we propose an automatic approach to synchronizing models that are related by model transformations. Given a unidirectional transformation between metamodels, we can automatically synchronize models in the metamodels by propagating modifications across the models. We have implemented a model synchronization system supporting the Atlas Transformation Language (ATL) and have successfully tested our implementation on several ATL transformation examples in the ATL web site.

Inferring Resource Specifications from Natural Language API Documentation

Typically, software libraries provide API documentation, through which developers can learn how to use libraries correctly. However, developers may still write code inconsistent with API documentation and thus introduce bugs, as existing research shows that many developers are reluctant to carefully read API documentation. To find those bugs, researchers have proposed various detection approaches based on known specifications. To mine specifications, many approaches have been proposed, and most of them rely on existing client code. Consequently, these mining approaches would fail to mine specifications when client code is not available. In this paper, we propose an approach, called Doc2Spec, that infers resource specifications from API documentation. For our approach, we implemented a tool and conducted an evaluation on Javadocs of five libraries. The results show that our approach infers various specifications with relatively high precisions, recalls, and F-scores. We further evaluated the usefulness of inferred specifications through detecting bugs in open source projects. The results show that specifications inferred by Doc2Spec are useful to detect real bugs in existing projects.

A propositional logic-based method for verification of feature models

The feature model is a domain/product-line oriented requirements model based on hierarchical structure and explicit variability modeling, and has been adopted by several important software reuse methods. However, with respect to the problem of verification of constraints on features and verification of partially customized feature models, these methods tend to be semi-formal and offer little formal assistance. In this paper, we propose a propositional logic-based method for the verification of feature models at different binding times. The most important characteristic of this method is that it integrates the logical verification with binding times, which makes it can be used to verify any partially customized feature models at any binding time (except run-time). In this method, constraints on features are formalized by logical sentences. Thus, the verification of feature models is converted into satisfaction problems in the logic. With this formal method, verification problems such as the detection of inconsistent constraints or the detection of conflicting or unnecessary binding resolutions can be automatically revealed.

Time-aware test-case prioritization using integer linear programming

Techniques for test-case prioritization re-order test cases to increase their rate of fault detection. When there is a fixed time budget that does not allow the execution of all the test cases, time-aware techniques for test-case prioritization may achieve a better rate of fault detection than traditional techniques for test-case prioritization. In this paper, we propose a novel approach to time-aware test-case prioritization using integer linear programming. To evaluate our approach, we performed experiments on two subject programs involving four techniques for our approach, two techniques for an approach to time-aware test-case prioritization based on genetic algorithms, and four traditional techniques for test-case prioritization. The empirical results indicate that two of our techniques outperform all the other techniques for the two subjects under the scenarios of both general and version-specific prioritization. The empirical results also indicate that some traditional techniques with lower analysis time cost for test-case prioritization may still perform competitively when the time budget is not quite tight.

A Runtime Model Based Monitoring Approach for Cloud

Monitoring plays a significant role in improving the quality of service in cloud computing. It helps clouds to scale resource utilization adaptively, to identify defects in services for service developers, and to discover usage patterns of numerous end users. However, due to the heterogeneity of components in clouds and the complexity arising from the wealth of runtime information, monitoring in clouds faces many new challenges. In this paper, we propose a runtime model for cloud monitoring (RMCM), which denotes an intuitive representation of a running cloud by focusing on common monitoring concerns. Raw monitoring data gathered by multiple monitoring techniques are organized by RMCM to present a more intuitive profile of a running cloud. We applied RMCM in the implementation of a flexible monitoring framework, which can achieve a balance between runtime overhead and monitoring capability via adaptive management of monitoring facilities. Our experience of utilizing the monitoring framework on a real cloud demonstrates the feasibility and effectiveness of our approach.

Discovering Homogeneous Web Service Community in the User-Centric Web Environment

The Web has undergone a tremendous change toward a highly user-centric environment. Millions of users can participate and collaborate for their own interests and benefits. Services computing paradigm together with the proliferation of Web services have created great potential opportunities for the users, also known as service consumers, to produce value-added services by means of service discovery and composition. In this paper, we propose an efficient approach to facilitating the service consumer on discovering Web services. First, we analyze the service discovery requirements from the service consumers perspective and outline a conceptual model of homogeneous Web service communities. The homogeneous service community contains two types of discovery: the search of similar operations and that of composible operations. Second, we describe a similarity measurement model for Web services by leveraging the metadata from WSDL, and design a graph-based algorithm to support both of the two discovery types. Finally, adopting the popular atom feeds, we design a prototype to facilitate the consumers to discover while subscribing Web services in an easy-of-use manner. With the experimental evaluation and prototype demonstration, our approach not only alleviates the consumers from time-consuming discovery tasks but also lowers their entry barrier in the user-centric Web environment.

Bridging the gap between the total and additional test-case prioritization strategies

In recent years, researchers have intensively investigated various topics in test-case prioritization, which aims to re-order test cases to increase the rate of fault detection during regression testing. The total and additional prioritization strategies, which prioritize based on total numbers of elements covered per test, and numbers of additional (not-yet-covered) elements covered per test, are two widely-adopted generic strategies used for such prioritization. This paper proposes a basic model and an extended model that unify the total strategy and the additional strategy. Our models yield a spectrum of generic strategies ranging between the total and additional strategies, depending on a parameter referred to as the p value. We also propose four heuristics to obtain differentiated p values for different methods under test. We performed an empirical study on 19 versions of four Java programs to explore our results. Our results demonstrate that wide ranges of strategies in our basic and extended models with uniform p values can significantly outperform both the total and additional strategies. In addition, our results also demonstrate that using differentiated p values for both the basic and extended models with method coverage can even outperform the additional strategy using statement coverage.

Is operator-based mutant selection superior to random mutant selection?

Due to the expensiveness of compiling and executing a large number of mutants, it is usually necessary to select a subset of mutants to substitute the whole set of generated mutants in mutation testing and analysis. Most existing research on mutant selection focused on operator-based mutant selection, i.e., determining a set of sufficient mutation operators and selecting mutants generated with only this set of mutation operators. Recently, researchers began to leverage statistical analysis to determine sufficient mutation operators using execution information of mutants. However, whether mutants selected with these sophisticated techniques are superior to randomly selected mutants remains an open question. In this paper, we empirically investigate this open question by comparing three representative operator-based mutant-selection techniques with two random techniques. Our empirical results show that operator-based mutant selection is not superior to random mutant selection. These results also indicate that random mutant selection can be a better choice and mutant selection on the basis of individual mutants is worthy of further investigation.

Runtime recovery and manipulation of software architecture of component-based systems

Recently, more attention is paid to the researches and practices on how to use software architecture in software maintenance and evolution to reduce their complexity and cost. The key in such architecture-based maintenance and evolution is to recover software architecture from existing systems. Almost all studies on architecture recovery focus on analyzing the source code and other documents. Such recovered software architecture can hardly capture runtime states and behaviors of the system. At the same time, current work pays little attention on how to change the system directly via manipulating the recovered software architecture. This paper presents a novel approach to recovering software architecture from component based systems at runtime and changing the runtime systems via manipulating the recovered software architecture. Containing much more details than the designed software architecture, the recovered software architecture can accurately and thoroughly describe the actual states and behaviors of the runtime system. It can be described formally with the extension of traditional architecture description language, which enables the recovered software architecture to absorb the semantics embedded in the designed software architecture. The recovered software architecture can be represented as multiple views so as to help different users to control the complexity from different concerns. Based on the reflective ability of the component framework, the recovered software architecture is up-to-date at any time and changes made on it will immediately lead to the corresponding changes in the runtime system. The approach presented in this paper is demonstrated on PKUAS, a reflective J2EE (Java 2 Platform Enterprise Edition) application server, and the performance is also evaluated.

Remote attestation on program execution

Remote attestation provides the basis for one platform to establish trusts on another. In this paper, we consider the problem of attesting the correctness of program executions. We propose to measure the target program and all the objects it depends on, with an assumption that the Secure Kernel and the Trusted Platform Module provide a secure execution environment through process separation. The attestation of the target program begins with a program analysis on the source code or the binary code in order to find out the relevant executables and data objects. Whenever such a data object is accessed or a relevant executable is invoked due to the execution of the target program, its state is measured for attestation. Our scheme not only testifies to a programs execution, but also supports fine-granularity attestations and information flow checking.