Electronic systems that use CPUs or processing platforms to perform aspecific function that is known up-front, are generally known as embeddedsystems. This is especially so when they are neither used nor perceived ascomputers [240]. Embedded systems are widely deployed in consumer productsthat we use everyday, such as TV sets, automobiles, mobile phones. Infact, International Data Corporation (IDC) reports that of the nearly 2 billionmicroprocessor chips manufactured each year, over 95% go into non-PC“embedded” devices.The ever-progressing semiconductor processing technique has dramaticallyincreased the number of transistors on a single chip, which makestoday’s embedded processors increasingly powerful and efficient. Sophisticatedmultiprocessor system-on-chips (MP-SoCs) have been used in thehigh-end embedded systems. On the other hand, the gap between the semiconductorprocessing capability and integrated circuit (IC) design capability isconstantly increasing, due to the faster growth of processing capability (58%per year) against the growth of design capability (21% per year). This designproductivity gap is increasing the design cost rapidly. Another problem thatthe designers will confront is the extremely high manufacturing non-recurringengineering (NRE) cost. As the semiconductor industry approaches the sub-100 nm technology node, the NRE (mask set and probe card) costs are gettingclose to $1 million for a large IC. With an average of just 500 wafersproduced from each mask set, rapid growth of manufacturing NRE can throttlethe initiation of new IC design projects. As a result of the high IC designand NRE cost, and also due to the stringent time-to-market deadlines, moreand more functionalities which used to be implemented in ASICs are migratingto embedded processors, and in general “processing platforms” [194].Conventional embedded system design techniques that rely on manual partitioningand scheduling of software components are clearly not able to meetthe requirements of today’s demanding software functionalities. The largeamount of software components which are present in embedded systemsrequire a novel high-level design methodology that can quickly explore theof Dynamic Concurrent Task-Based Systems on Heterogeneous Platformsdesign space and hence map them onto the multiple embedded processorsin an efficient manner.In this chapter, we first give a brief des cription of the multiprocessorSoC hardware platforms (Section 1.1). Then we identify the characteristicsof the demanding target applications (Section 1.2). After that, we discussthe still missing design techniques to efficiently map the current andfuture applications on the target hardware platforms, and hence introduceour Task Concurrency Management (TCM) flow, a high-level embeddedsoftware design methodology, to facilitate the embedded system design(Section 1.3). We also provide already a brief related work comparisonthen. Subsequently, we present the essential structure of the existing TCMwork flow (Section 1.4). Finally, we present a chapter overview of this book(Section 1.5).
猜您喜欢
评论