OPTIMIZING RESOURCE UTILIZATION USING TRANSFORMATIONS

The goal of the high level synthesis process for real time applications is to minimize the implementation cost, while still satisfying all timing constraints. In this paper, we present how a combination of four conceptually simple, yet powerful, transformations: namely retiming, associativity, commutativity and inverse element law, can help to further this goal. Since the minimization problem associated with those transformations is NP complete, a new fast iterative improvement probabilistic algorithm has been developed. The effectiveness of the proposed algorithm and the associated transformations is demonstrated in multiple ways: using standard benchmark examples, with the aid of statistical analysis and through a comparison with estimated minimal bounds.