MacroRank: Ranking Macro Placement Solutions Leveraging Translation Equivariancy

• Main Authors: Chen, Yifan, Mai, Jing, Gao, Xiaohan, Zhang, Muhan, Lin, Yibo
• Format: Conference Proceeding
• Language: English
• Subjects: Applied computing; Applied computing > Arts and humanities; Applied computing > Arts and humanities > Architecture (buildings); Applied computing > Arts and humanities > Architecture (buildings) > Computer-aided design; Applied computing > Physical sciences and engineering; Asia; Benchmark testing; Correlation coefficient; Design automation; Hardware; Hardware > Electronic design automation; Hardware > Electronic design automation > Physical design (EDA); Hardware > Electronic design automation > Physical design (EDA) > Placement; Hardware > Electronic design automation > Physical design (EDA) > Wire routing; Hardware > Hardware validation; Measurement; Predictive models; Routing
• ISSN: 2153-697X
• DOI: 10.1145/3566097.3567899

Modern large-scale designs make extensive use of heterogeneous macros, which can significantly affect routability. Predicting the final routing quality in the early macro placement stage can filter out poor solutions and speed up design closure. By observing that routing is correlated with the relative positions between instances, we propose MacroRank, a macro placement ranking framework leveraging translation equivariance and a Learning to Rank technique. The framework is able to learn the relative order of macro placement solutions and rank them based on routing quality metrics like wirelength, number of vias, and number of shorts. The experimental results show that compared with the most recent baseline, our framework can improve the Kendall rank correlation coefficient by 49.5% and the average performance of top-30 prediction by 8.1%, 2.3%, and 10.6% on wirelength, vias, and shorts, respectively.