EVALUATING MULTIGAUGE ARCHITECTURES FOR COMPUTER VISION

Heterogeneous computing is a technique for achieving high performance by providing a variety of different architectures to meet the needs of systems that are composed of tasks with widely different characterist ics. Essential to the construction of heterogeneous systems is an unde rstanding of the match between architecture and software and how that match can be used in deciding how to utilize the available computing r esources. We present a theoretical framework, the PCI model, which def ines corresponding characteristics of parallel programs and parallel a rchitectures and defines the performance relationship between them in terms of these characteristics. We have encapsulated the concepts of t he PCI model into RAW, a simulation environment that facilitates exper imentation with the program/architecture relationship in terms of the PCI model. Using RAW, we have applied the PCI model to study the use o f processor reconfigurable architectures (a type of heterogeneous syst em) in the context of computer vision applications. We present experim ental results that demonstrate that these architectures perform better than static homogeneous architectures for such applications. (C) 1994 Academic Press, Inc.