MADNESS · Methods for predictAble Design of heterogeNeous Embedded System with adaptivity and reliability Support

The MADNESS project aims at the definition of innovative system-level design methodologies for embedded systems, able to drive the optimal composition of an heterogeneous MPSoC architecture, according to the requirements and the features of a given target application field.<br/>The proposed methodologies will extend the classic concept of design space exploration, to cope with high heterogeneity, technology scaling, system reliability and multi-application demands, pursuing the following objectives:<br/>-Improve design predictability of highly heterogeneous embedded systems, bridging the so called implementation gap", i.e. the gap between the results that can be predicted during the system-level design phase and those eventually obtained after the on-silicon implementation.<br/>-Consider, apart from more traditional constraints (typically, cost, performance, power consumption), continued availability of service, taking into account fault recovery as one of the optimization factors to be satisfied.<br/>-Support adaptive runtime management of the architecture.<br/>The technical approach of the project will rely on the following methods:<br/>-In order to improve the design predictability, traditional system-level design methodologies will be extended to consider variables strictly related with physical implementation of the architecture, leveraging a specific layer for rapid and accurate on-hardware FPGA-based emulation.<br/>-In order to address fault tolerance and adaptive runtime resources management, new methodologies are going to be defined and included in system-level exploration. Specific hardware/middleware IP modules will be developed to pursue those objectives.<br/>The results of the project are expected to have a deep impact on the design flow for high-complexity embedded systems, facilitating and enhancing the exploration of the architectural design space, therefore resulting in a significantly increased overall productivity."