|8:15 -9:30||Opening and Keynote by Rajeev Alur (Common to all workshops)|
|9:30 – 10:00||Networking Break|
|10:00 – 10:45||Invited keynote by Dr. Henry Hoffmann|
|10:45 – 11:15||Invited talk by Prof. Rolf Ernst
- "Controlling Concurrent Change - A self-aware infrastructure for continuous change and evolution in automotive systems"
|11:15 - 12:00||
Presentations from participants, based on the CfP
- "Attaining self-awareness in Cyber-Physical Systems", L. Mutos, J. Preden, J. Ehala
- "Towards Dynamic Architectures for Computational Self-Awareness", P. Lewis
- "Self-aware Internet-of-Things Based Medical Early Warning Score System", A. Anzanpour, I. Azimi, A. M. Rahmani, P. Liljeberg, H. Tenhunen
|13:30 -15:00||Hands-on Activity, Part I: Testing specific systems for self-awareness properties|
|15:30 – 17:00||Hands-on Activity, Part II: Testing specific systems for self-awareness properties|
|17:00 – 18:00||Panel Discussion|
Modern computing applications have to meet multiple -- often competing -- goals; for example providing high performance and low energy consumption. At the same time, the computing platforms used to implement these applications are becoming increasingly complex, integrating multiple cores of different types and capabilities. Meeting a diverse set of application requirements on complicated hardware platforms requires developers to be experts in both their application domain and myriad systems issues concerning power, performance, and other tradeoffs. This is simply an unrealistic burden for most programmers.
To alleviate this burden, we propose a new, self-aware computational model, SEEC. Using SEEC, programmers specify high-level goals and optimization criteria as well as the changeable components of the system and application. SEEC's runtime then uses a combination of machine learning (for optimization) and control theory (for goal management) to ensure that users’ goals are met while the remaining behavior is optimized. In this talk, we provide an overview of SEEC, compare it to other approaches that have been called self-aware, and describe some cases studies showing the benefits of self-awareness. We then list some of the challenges of applying this type of computing model to cyber-physical systems.
Henry Hoffmann has been an Assistant Professor in the Department of Computer Science at the University of Chicago since January 2013 where he leads the Self-aware computing group (or SEEC project) and conducts research on adaptive techniques for power, energy, and performance management in computing systems.
He has spent the last 13 years working on multicore architectures and system software in both academia and industry. He completed a PhD in Electrical Engineering and Computer Science at MIT where his research on self-aware computing was named one of the ten "World Changing Ideas" by Scientific American in December 2011. He received his SM degree in Electrical Engineering and Computer Science from MIT in 2003. As a Masters student he worked on MIT's Raw processor, one of the first multicores. Along with other members of the Raw team, he spent several years at Tilera Corporation, a startup which commercialized the Raw architecture and created one of the first manycores. His implementation of the BDTI Communications Benchmark (OFDM) on Tilera's 64-core TILE64 processor still has the highest certified performance of any programmable processor.
Prior to his graduate studies, he served as an Associate Staff member of MIT Lincoln Laboratory where his research produced the Parallel Vector Library (PVL), which forms the foundation of the VSIPL++ standard, an interface for parallel signal and image processing. Henry was appointed as a Lincoln Masters Scholar in 2001 and a Lincoln Doctoral Scholar in 2004. In 1999, he received his BS in Mathematical Sciences with highest honors and highest distinction from UNC Chapel Hill.
© 2016 TU Vienna - Institute of Computer Technology, Inc.