<inline-formula><tex-math notation="LaTeX">${\mathsf{E-MGN}}$</tex-math><alternatives><mml:math><mml:mrow><mml:mi mathvariant="sans-serif">E</mml:mi><mml:mo>-</mml:mo><mml:mi mathvariant="sans-serif">MGN</mml:mi></mml:mrow></mml:math><inline-graphic xlink:href="rao-ieq1-3746439.gif"/></alternatives>
Abstract
Numerical simulation is important in many engineering domains such as aerodynamics and structural mechanics. Researchers have developed message passing (MP)-based multilevel mesh graph networks (MGNs) for numerical simulation with fast prediction time and acceptable accuracy. Unfortunately, when the simulation (e.g., the numerical simulation of air pressure on a car surface within a wind tunnel) involves the complex interactions between simulation objects (car surface) and external conditions (the initial air velocity), existing MP schemes in MGNs face effectiveness and efficiency issues in modeling such simulations. In this paper, we propose a scalable MGN model with a novel edge-centric message passing scheme on large mesh graphs, namely <inline-formula><tex-math notation="LaTeX">${\mathsf{E-MGN}}$</tex-math><alternatives><mml:math><mml:mrow><mml:mi mathvariant="sans-serif">E</mml:mi><mml:mo>-</mml:mo><mml:mi mathvariant="sans-serif">MGN</mml:mi></mml:mrow></mml:math><inline-graphic xlink:href="rao-ieq3-3746439.gif"/></alternatives></inline-formula>. Firstly, the edge-centric MP scheme (resp. coarse-mesh to fine-mesh refinement) can effectively capture local (resp. global) interactions. Next, for high efficiency, <inline-formula><tex-math notation="LaTeX">${\mathsf{E-MGN}}$</tex-math><alternatives><mml:math><mml:mrow><mml:mi mathvariant="sans-serif">E</mml:mi><mml:mo>-</mml:mo><mml:mi mathvariant="sans-serif">MGN</mml:mi></mml:mrow></mml:math><inline-graphic xlink:href="rao-ieq4-3746439.gif"/></alternatives></inline-formula> performs the MP in parallel on distributed mesh partitions with low communication overhead. Evaluation on three datasets (two synthetic and one real) against previous baseline algorithms demonstrates the superiority of <inline-formula><tex-math notation="LaTeX">${\mathsf{E-MGN}}$</tex-math><alternatives><mml:math><mml:mrow><mml:mi mathvariant="sans-serif">E</mml:mi><mml:mo>-</mml:mo><mml:mi mathvariant="sans-serif">MGN</mml:mi></mml:mrow></mml:math><inline-graphic xlink:href="rao-ieq5-3746439.gif"/></alternatives></inline-formula>. For example, on the <italic>Beam</italic> dataset, <inline-formula><tex-math notation="LaTeX">${\mathsf{E-MGN}}$</tex-math><alternatives><mml:math><mml:mrow><mml:mi mathvariant="sans-serif">E</mml:mi><mml:mo>-</mml:mo><mml:mi mathvariant="sans-serif">MGN</mml:mi></mml:mrow></mml:math><inline-graphic xlink:href="rao-ieq6-3746439.gif"/></alternatives></inline-formula> leads to 6.95% smaller RMSE than the state-of-the-art and leads to 6.93× speedup on 8 GPUs.