Thomas Gleixner, Real-Time Linux Kernel Maintainer
The Realtime Preemption Patch: Pragmatic Ignorance or a Chance to
text available here
The realtime preemption patch is a project that aims to provide real-time capabilities to the Linux kernel. Like many other open source projects, it follows a pragmatic approach, which has been criticized more than once for ignoring academic research results.
This raises a few questions:
- Is a pragmatic approach inevitably ignorant?
- Is academic research inevitably non-pragmatic?
- Is there a way to connect both worlds with mutual benefits?
There are no easy answers to these questions, but ignoring them deliberately would be the least pragmatic approach.
Thomas Gleixner has worked on industrial embedded devices for more
than 20 years. He's CTO of Linutronix, a Germany based Linux
consulting and service provider.
During the last 10 years Thomas has contributed to the Linux kernel
in significant ways. He (co)maintains several core areas in the
kernel (timer subsystem, interrupt subsystem, x86 architecture).
Aside from that, he maintains the realtime preemption patch, which
aims to provide real-time capabilities to the Linux kernel.
Norbert Wehn, Technische Universität Kaiserslautern, Germany, Chairman, European Design and Automation Association, DATE Conference
Hardware Modeling: A Critical Assessment with Case Studies
slides available here
Abstraction and the use of models play an essential role in embedded systems design, as they allow to deal with the vast complexity of today's systems and applications. It is crucial to hide the details of lower design levels, like complex transistor behavior, from higher layers. This is achieved by identifying and focusing solely on those aspects that expose relevant higher-level effects. However, many existing models are overly simplified and not practically evaluated. With the advance in microelectronics and architectures this problem becomes even worse.
In this talk I want to assess some common models and beliefs. I will put special emphasis on models for power and energy consumption, as power optimization is
one of the key challenges in embedded systems design. I will show case studies to prove that many of these models promote misleading optimization strategies.
Norbert Wehn holds the chair for Microlectronic Systems Design in the department of Electrical Engineering and Information Technology at the University of Kaiserslautern. He has published more than 200 articles in conferences and journals in various fields of microelectronic systems design and holds several patents. He is chairman of the European Design Automation Association, chairman of the Research Center "Ambient Systems" TU Kaiserslautern, associate editor of various journals and a member of several scientific advisory boards. In 2003 he served as program chair for the DATE 2003 conference and as general chair for DATE 2005, respectively. His special research interests are VLSI-achitectures for mobile communication, forward error correction techniques, low-power, advanced SoC architectures, and reliability issues in SoC.
Sanjoy Baruah, The University of North Carolina, Chapel Hill, USA
Why Real-Time Scheduling Theory Still Matters
slides available here
The relationship between real-time scheduling theory and the broader discipline of real-time systems has of late become a somewhat uneasy one: there is concern that real-time scheduling is becoming "too theoretical," and is focusing on problems of little relevance to the actual issues confronting designers and builders of real-time systems. This presentation will revisit this relationship, highlighting past successes of real-time scheduling theory in assisting systems designers and seeking to learn lessons from these past successes in order to propose a more relevant and useful agenda for real-time scheduling research.
Sanjoy Baruah is a professor in the Department of Computer Science at the University of North Carolina at Chapel Hill. He received his Ph.D. from the University of Texas at Austin in 1993. His research and teaching interests are in scheduling theory, real-time and safety-critical system design, and resource-allocation and sharing in distributed computing environments.