Renato Mancuso is a doctoral candidate in the Department of Computer
Science at the University of Illinois at Urbana-Champaign. He is
interested in high-performance cyber-physical systems, with a specific
focus on techniques to enforce strong performance isolation and
temporal predictability in multi-core systems. He has published around
20 papers in major conferences and journals. His papers were awarded a
best student paper award and a best presentation award at the
Real-Time and Embedded Technology and Applications Symposium (RTAS) in
2013 and 2016, respectively. He was the recipient of a Computer
Science Excellence Fellowship, and a finalist for the Qualcomm
Innovation Fellowship. Some of the design principles for real-time
multi-core computing proposed in his research have been officially
incorporated in recent certification guidelines for avionics
systems. They have also been endorsed by government agencies,
industries and research institutions worldwide. He received a B.S. in
Computer Engineering with honors (2009) and a M.S. in Computer
Engineering with honors (2012) from the University of Rome "Tor
Vergata".

AG 1, AG 2, AG 3, AG 4, AG 5, SWS, RG1, MMCI   Info

AG Audience

English

-- Not specified --

Kaiserslautern

There has been an uptrend in the demand and need for complex Cyber-Physical Systems (CPS), such as self-driving cars, unmanned aerial vehicles (UAVs), and smart manufacturing systems for Industry 4.0. CPS  often need to accurately sense the surrounding environment by using high-bandwidth acoustic, imaging and other types of sensors; and to take coordinated decisions and issue time critical actuation commands. Hence, temporal predictability in sensing, communication, computation, and actuation is a fundamental attribute. Additionally, CPS must operate safely even in spite of software and hardware misbehavior to avoid catastrophic failures. To satisfy the increasing demand for performance, modern computing platforms have substantially increased in complexity; for instance, multi-core systems are now mainstream, and partially re-programmable system-on-chip (SoC) have just entered production. Unfortunately, extensive and unregulated sharing of hardware resources directly undermines the ability of guaranteeing strong temporal determinism in modern computing platforms. Novel software architectures are needed to restore temporal correctness of complex CPS when using these platforms. My research vision is to design and implement software architectures that can serve as a reference for the development of high-performance CPS, and that embody two main requirements: temporal predictability and robustness. In this talk, I will address the following questions concerning modern multi-core systems: Why application timing can be highly unpredictable? What techniques can be used to enforce safe temporal behaviors on multi-core platforms? I will also illustrate possible approaches for time-aware fault tolerance to maximize CPS functional safety. Finally, I will review the challenges faced by the embedded industry when trying to adopt emerging computing platforms, and I will highlight some novel directions that can be followed to accomplish my research vision.

Roslyn Stricker

--email hidden

Yes

Saarbrücken

E1 5

029

passcode not visible

logged in users only