Events login

Research and Discovery | Research Events

Warning Icon This event is in the past.
September 18, 2018 | 11:30 a.m. - 12:20 p.m.
Category: Seminar
Location: State Hall #234 | Map
5143 Cass
Detroit, MI 48202
Cost: Free
Audience: Academic Staff, Alumni, Community, Current Graduate Students, Current Undergraduate Students, Faculty, Parents, Prospective Students, Staff

Energy-Efficient Mobile Computing

Abstract: Two critical quality factors for mobile devices (e.g., smartphones, tablets) are battery life and apps’ user perceived performance. For example, apps that require frequent user actions with the user interface should have high responsiveness, which indicates how fast an app reacts to user actions. On the other hand, apps used mostly for video/music play should have a high throughput, which allows for example a video to be played smoothly without perceivable frame rate loss. Two main causes of poor user-perceived performance for these apps are soft hang bugs (i.e., programming bugs that can cause apps to have perceivable delays) and resource contention among concurrently executing foreground apps (e.g., watch a video while chatting with a friend), which may lead to performance imbalance and thus poor performance or high energy consumption.

In this seminar, I first present two runtime methodologies: Hang Doctor and SURF. Hang Doctor efficiently detects and diagnoses soft hang bugs in mobile apps. It exploits performance event counters to ensure high detection quality and low overhead. We have implemented Hang Doctor and tested it with the latest releases of 114 real-world apps and found 34 new soft hang bugs previously unknown to their developers. SURF is a runtime resource management algorithm that first dynamically manipulates the app priorities to balance the app performance, and second, exploits supervisory control theory to efficiently handle aperiodic user actions. We test SURF on several mobile device models with real-world open-source apps and show that it can reduce the CPU energy consumption by 30-90% compared to state-of-the-art solutions while causing no perceivable performance degradation. Then, I present an example of my future research in energy-efficient mobile computing. I particular, I present how a collection of smart devices near each other could be exploited to reduce and redistribute the workload executed by each device for lower (overall) energy consumption.

Bio: Marco Brocanelli is an assistant professor in the computer science department of Wayne State University. He got his PhD degree in the Department of Electrical and Computer Engineering at The Ohio State University,  in 2018. He received an MS degree in Control Engineering at University of Rome Tor Vergata, Italy, in 2011. He received a BS degree in Control Engineering at University of Rome Tor Vergata, Italy, in 2008. He was a J1 visiting scholar at The Ohio State University between October 2010 and April 2011 - October 2011 and May 2012, working on Hypersonic Vehicle non-linear Control. His research interests are Cyber-physical systems, Energy-aware systems, Internet-of-Things, Edge Computing, Embedded and real time systems.

For more information about this event, please contact LaNita Stewart at 313-577-2478 or LStewart@wayne.edu.