On the optimization of regular wavefront arrays

The author examines systematic methods for analyzing and designing self-timed regular arrays of processors. Methods are given to derive measures of efficiency and to verify the computational behavior of a given self-timed regular array. It is shown that the optimization of a given regular self-timed processor array with respect to its processor utilization can be given mathematically in the form of linear programming or integer linear programming problems whose sizes are independent of the size of the array.<
>