Thermal-aware rotation scheduling for 3D multi-core with timing constraint

Multi-core technique has been implemented in DSP systems due to the tremendous computation requirements. In order to integrate more functionalities and provide higher performance, 3D multi-core architecture has been studied recently. The thermal issue is a critical problem for 3D architecture. In this paper, we propose an online thermal prediction model for 3D chip. Using this model, we present a task scheduling algorithm based on rotation scheduling to reduce the peak temperature on chip. While existing works do not consider the inter-iteration data dependencies in the application, we consider these dependencies in our proposed algorithm. Our simulation result shows that our algorithm can efficiently reduce the peak temperature up to 8°C.