arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2211.14969math.NAcs.NA
收藏

二维分层庞加莱-斯捷克洛夫格式的GPU优化

GPU Optimizations for the Hierarchical Poincaré-Steklov Scheme

Anna Yesypenko, Per-Gunnar Martinsson

AI总结:

针对二维分层庞加莱-斯捷克洛夫(HPS)离散格式提出GPU优化方法,通过利用混合架构批量线性代数降低高局部阶p离散的局部静态凝聚成本,提升求解器速度,实验验证其高精度与计算效率提升。

摘要:

本文提出了针对二维分层庞加莱-斯捷克洛夫(HPS)离散格式的GPU优化方法。HPS是一种多域谱配置方法,结合高阶离散与直接求解器,可精确求解高度振荡解。HPS的区域分解方法通过稀疏直接求解器直接连接域。所提优化利用现代混合架构上的批量线性代数,实现简单,提升了求解器实际速度。研究表明,GPU优化可显著降低高局部阶p离散中传统局部静态凝聚的高昂成本。对高波数亥姆霍兹方程在曲面和矩形域的数值实验,验证了HPS离散的高精度及GPU优化带来的计算时间显著减少。

英文摘要:

This manuscript presents GPU optimizations for the 2D Hierarchical Poincaré-Steklov (HPS) discretization scheme. HPS is a multi-domain spectral collocation method that combines high-order discretizations with direct solvers to accurately resolve highly oscillatory solutions. The domain decomposition approach of HPS connects domains directly via a sparse direct solver. The proposed optimizations exploit batched linear algebra on modern hybrid architectures, are straightforward to implement, and improve the solver's practical speed. The manuscript demonstrates that GPU optimizations can significantly reduce the traditionally high cost of performing local static condensation for discretizations with very high local order $p$. Numerical experiments for the Helmholtz equation with high wavenumbers on curved and rectangular domains confirm the high accuracy achieved by the HPS discretization and the significant reduction in computation time achieved with GPU optimizations.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑