ABSTRACT

This chapter introduces the concept of Algorithm/Architecture Co-Exploration (AAC), which is a leading paradigm. It provides quantitative measures of the intrinsic complexity of algorithms for envisioning even the scope or spectrum of future system architectures and platforms. The chapter examines the concept of AAC so as to increase the values of the visual computing systems including flexibility, scalability, gate count, power consumption, maintainability, design turn-around time, and the risk of failure. It describes the importance of concurrently optimizing both algorithm and architecture so that architecture information is back-annotated to the algorithm and is thus being considered as early as the algorithm design phase. Computer graphics algorithms are highly computationally intensive: some graphics algorithms, such as real-time rendering, demand a huge amount of processing power, and visualizing graphic content is possible when using many tremendously complex graphics algorithms. The complexity metrics have to be intrinsic to the algorithm and hence are not biased toward either hardware or software.