Toward a Practical Methodology for Completely Characterizing the Optimal Design Space

One of the most compelling reasons for developing highlevel synthesis systems has been the desire to quickly explore the design space. Since this problem is very difficult to solve optimally, most systems compute either lower bounds or estimates on the optimal tradeoff curve. The methodology described here goes beyond most previous work in several ways: (1) it computes all optimal tradeoff points so as to completely characterize the design space, (2) it solves not only the scheduling problem, but the clock determination and module selection problems as well, and (3) it carefully prunes the search space at each level of the design cycle. 1 Introduction For many years, one of the most compelling reasons for developing high-level synthesis systems [1, 2] has been the desire to quickly explore a wide range of designs for the same behavioral description. Given a set of designs, two metrics are commonly used to evaluate their quality: area (ideally total area, but often only functional uni..