The coming of age of reconfigurable computing-potentials and challenges of a new technology [Tutorial]

Addresses modelling for analysis and optimisation of MPSoC architectures including hardware and software layers such as drivers, run-time systems and application APIs. Different scheduling strategies, communication behaviours and performance requirements must be matched and combined on a single system. 4. Presents the DELI experience to show that application level software development can be overlapped with the development of the rest of the system. DELI is aimed to generating portable HdS taking into account real-time constraints and low-level interactions with OS and the HW during early prototyping. Platforms that combine CPUs with a reconfigurable fabric on the same chip have been recently introduced. Such devices are ideally suited for many application domains ranging from multimedia to communication. A major challenge to their wider use is the lack of high level programming and design space exploration tools. This tutorial focuses on three aspects of this emerging technology: (1) the intrinsic potential of the temporal/spatial paradigm (2) The wide range of architectures, fine and coarse grained, and the trade-offs between performance and flexibility. (3) A survey of current and future applications in multimedia and mobile communication and analysis of their performance and energy requirements. It is intended for an audience of developers and researchers in high-end embedded systems. The tutorial presents methods for reliable system design. We consider systems realised on a single chip, systems consisting of several integrated components (possibly components off-the shelf-COTS), and distributed systems. We address both hardware (computation and communication) and software aspects of reliable system design. This tutorial is intended for researchers in EDA, system/chip designers and software developers for integrated systems, as well as for managers who want to learn about reliable design and how this area evolved in view of the current technology trends. The tutorial covers a spectrum of diagnosis and debug topics from the basic concepts to future challenges, over the life-cycle of a product. Established diagnosis procedures fault dictionaries, post-test fault simulation, and hardware-based backtracking will be discussed, followed by recent enhancements and advanced diagnosis topics, including methods for locating defects, approximation techniques for identifying unmodelled faults, deductive analysis, Iddq-based diagnosis, diagnosis for delay-faults, scan-chain diagnosis, BIST-based diagnosis, and design-for-diagnosability techniques. The tutorial then will focus on silicon debug techniques, design-for-debug techniques, and applications to yield improvement. Successful diagnosis methods used in real industrial products, industrial experiences, and case studies will be presented in this tutorial. determined to find out where those puzzling …