Proactive performance assurance in Mobeewave****

点击翻译按钮获取中文摘要

Mobeewave offers their award-winning services to tackle the challenges of mobile payment with their software solutions of in-person peer-to-peer and micro-merchant transactions. With their software solutions, it possible for anyone to accept money using their phones' NFC capability. Due to the strong need of Mobeewave's software solution, and the criticality of mobile payment solution, performance has become an detrimental factor to the success of Mobeewave. Our goal is to to design automated approaches that can proactively predict and detect performance issues. The proposed work intersects two areas of software engineering: software performance and software quality. Prior research proposes to monitor and improve software performance with various automated techniques. However, identifying performance issues remains a task that is conducted after fact, i.e., after the system is built and deployed in the field or dedicated performance testing environments. Large amounts of resources are required to detect, locate, understand and fix performance regressions at such a late stage in the development circle. We plan to leverage techniques that are based on machine learning and mining software repositories that provides just-in-time prediction and localization on the software changes that may introduce negative impact on system performance. The outcome of this work will advance the state- of-the-art in performance assurance activities in the area of software engineering in general and provide Mobeewave with an advanced infrastructure to support their performance assurance activities.**