Architecture-Based Software Reliability Estimation: Problem Space, Challenges, and Strategies

Student: Ivo Krka, USC

Collaborators: Leslie Cheung, USC, George Edwards, USC, Leana Golubchik, USC

Advisor: Nenad Medvidovic, USC/ISR

Abstract:

In modern software-intensive systems, reliability is considered to be one of the most critical non-functional properties. To build software in a cost-efficient manner, reliability should be analyzed at ar-chitecture design time. Firstly, we present the problem space of architecture-based estimation of a software system's reliability. Architecture-based reliability estimation is challenging — during early design phases, architects lack information necessary for precisely determining a system's reliability, such as the system's operational profile and failure and recovery information. Thus, we explore how such information can be obtained from alternative sources. Finally, we present a critical overview of existing approaches to architecture-based reliability estimation, in order to indicate directions for future research.

Bio:

Ivo Krka is a Ph.D. student in the Computer Science Department at the University of Southern California, where he is a member of the Software Architecture Research group, advised by Dr. Nenad Medvidovic. Additionally, Ivo is a recepient of USC Provost's Fellowship. His main research interests are architecture-based performance and reliability estimation. Before coming to USC, Ivo was a student at Faculty of Electrical Engineering and Computing (FER), University of Zagreb, Croatia, where he obtained M.Eng. (dipl.ing.) in Computing degree.