Exploring embedded-systems architectures with artemis

Because embedded systems mostly target mass production and often run on bat teries, they should be cheap to realize and power efficient. In addition, t hey require a high degree of programmability to provide real-time performan ce for multiple applications and standards. However, performance requirements as well as cost and power-consumption con straints demand that substantial parts of these systems be implemented in d edicated hardware blocks. As a result, their heterogeneous system architect ure consists of components ranging from fully programmable processor cores to fully dedicated hardware components for time-critical application tasks. Increasingly, these designs yield heterogeneous embedded multiprocessor sy stems that reside together on a single chip. The heterogeneity of these hig hly programmable systems and the varying demands of their target applicatio ns greatly complicate system design. The increasing complexity of embedded-system architectures makes predicting performance behavior more difficult. Therefore, having the appropriate too ls to explore different choices at an early design stage is increasingly im portant. The Artemis modeling and simulation environment aims to efficientl y explore the design space of heterogeneous embedded-systems architectures at multiple abstraction levels and for a wide range of applications targeti ng these architectures. The authors describe their application of this meth odology in two studies that showed promising results, providing useful feed back on a wide range of design decisions involving the architectures for th e two applications.