Neural Network-Based Thermal Simulation of Integrated Circuits on GPUs

With the rising challenges in heat removal in integrated circuits (ICs), the development of thermal-aware computing architectures and run-time management systems have become indispensable to the continuation of IC design scaling. These thermal-aware design technologies of the future strongly depend on the availability of efficient and accurate means for thermal modeling and analysis. These thermal models must have not only the sufficient accuracy to capture the complex mechanisms that regulate thermal diffusion in ICs, but also a level of abstraction that allows for their fast execution for design space exploration. In this paper, we propose an innovative thermal modeling approach for full-chips that can handle the scalability problem of transient heat flow simulation in large 2D/3D multi-processor ICs. This is achieved by parallelizing the computation-intensive task of transient temperature tracking using neural networks and exploiting the computational power of massively parallel graphics processing units (GPUs). Our results show up to 35x run-time speed-up compared to state-of-the-art IC thermal simulation tools while keeping the error lower than 1ºC. Speed-ups scale with the size of the 3D multi-processor ICs and our proposed method serves as a valuable design space exploration tool.