A Survey Based on Performance-Bug Detection

Software testing is the process of executing a program or system to discover the intent of bugs. Software testing is an activity designed to assess the properties or capabilities of a program or system and to determine the results that meet its requirements. Software performance is critical to how the user perceives the quality of the software product. Performance bugs are programming bugs that cause significant performance degradation, resulting in poor user experience and low system performance bugs. Designing effective techniques for handling bugs in yield design require a thorough understanding of how bugs are detected, reported and fixed. In this paper, to study how performance bugs, reported by developers to discover and develop fixation, and the results compared with non performance bugs. This paper conducts a detailed study of a sample of real-world performance bugs randomly from three sets of representative programs (Chrome, Mozilla and Apache). The results of this study serve as a guide for future work to prevent, expose, discover and correct performance bugs. First, found that there is little evidence that bug fixes are more likely to introduce new functional bugs to correct for erratic irregularities, which means that developers may not need more attention to correcting the performance of the bugs. Second, although fixing performance bugs is about as bug-prone as fixing nonperformance bugs, fixing performance bugs is more difficult than fixing non-performance bugs, indicating that developers need better tool support for fixing performance bugs and testing performance bug patches. Third, unlike many insects, a large percentage of performance bugs are detected by reasoning code, not by the user observing the negative effects of bugs or via profiles. The results show that the techniques that help developers need to test performance, better test beds and better profiling techniques to find performance bugs.