CAREER: Leveraging Field Data to Test Highly-Configurable and Rapidly-Evolving Pervasive Systems

University Of Nebraska-Lincoln, Lincoln NE

Investigators

Abstract

ABSTRACT 0347518 Sebastian Elbaum University of Nebraska-Lincoln CAREER: Leveraging Field Data to Test Highly-Configurable and Rapidly Evolving Pervasive Systems Software test engineers cannot predict, much less exercise, the overwhelming number of potential scenarios faced by their software. Instead, they allocate their limited resources based on assumptions about how the software will be employed after release. Yet, the lack of connection between in-house activities and how the software is employed in the field can lead to inaccurate assumptions, resulting in decreased software quality and reliability over the system's lifetime. This project will address those challenges by developing profiling techniques with a small footprint to capture field data and adaptive testing techniques that can leverage that information. The two primary objectives are: 1) to design profiling mechanisms for large pools of users and resource-cognizant monitoring schemes to reduce execution overhead, and 2) to develop adaptive testing techniques that continuously characterize software behavior to identify testing inadequacies, realign testing resources, and extend existing test suites to overcome their limitations. Empirical studies will be regularly employed for technique evaluation and exploration. The educational component of this work aims at bringing an earlier appreciation of the real-world validation problems to the undergraduate classroom and prepare students well-versed in software quality and empirical validation.