DEMO: Multi-Grain Adaptivity in Cyber-Physical Systems

Abstract

Cyber-Physical Systems (CPS) operate in increasingly complex and demanding application scenarios, while requiring also high adaptivity levels to satisfy several requirements that usually change over time. Multi-grain reconfigurable hardware architectures are an appealing solution to provide high, heterogeneous flexibility, thus reaching the advanced runtime adaptivity support necessary for CPS. This work presents a demonstration of an automated framework for the development and runtime management of multi-grain reconfigurable hardware systems. The framework supports different reconfiguration mechanisms, each with different overheads and depth in terms of system behavior modification.