Evaluation of routability-driven macro placement with machine-learning technique

Macro placement is an important step in floor-plan. Macros' location directly affects the next steps, cells placement and wires routing. However, it is a time-consuming work to evaluate macro placement's result. To address this problem, we propose an effective evaluation method with machine learning technique. Our methodology predicts HPWL and routing congestion after macro placement, rather than after cells placement and global routing. Therefore, it takes much short time that we can get HPWL and routing congestion. Experiment results show that our evaluation is accurate and effective.