An Optimal Lower-Bound Algorithm for the High-Level Synthesis Scheduling Problem

An optimal scheduling algorithm in high-level synthesis is described, which searches for the minimum time scheduling under given resource constraints. The algorithm can substantially reduce the computational effort required to obtain the exact solution using a branch-and-bound search method and some novel bounding techniques to prune the unfruitful search space. Some time properties are explored to enlighten these bounding techniques. Experimental results on several benchmarks with varying resource constraints show our algorithm is both efficient and effective.