Power and thermal constrained test scheduling

Abstract

We propose a test scheduling algorithm that ensures the resource compatibility and satisfies both power and thermal constraints. The proposed algorithm can start a test at an arbitrary time and it has the capability of delaying a test to let a core cool down to find a valid schedule even when traditional scheduling schemes cannot find a solution. To reduce the execution time of thermal simulation, we exploit superposition principle to compute the thermal profile rapidly and accurately. We apply our scheduling algorithm to ITC'02 SoC benchmarks and the results show a remarkable improvement in the total test length over other methods, while meeting the thermal and power constraints.