arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2512.08571physics.comp-ph
收藏

基于有限元离散化的分布式CPU架构中快速进行从头计算的无矩阵算法

Matrix-free algorithms for fast ab initio calculations on distributed CPU architectures using finite-element discretization

Indian Institute of Science(印度科学学院)

Gourab Panigrahi, Phani Motamarri

AI总结:

本研究提出无矩阵算法,用于在分布式CPU架构上加速有限元离散化的DFT计算,通过优化数据布局和算术强度,实现显著的速度提升。

摘要:

有限元(FE)离散化已作为一种强大的实空间替代方法,用于大规模Kohn-Sham密度泛函理论(DFT)计算,提供了系统收敛性、优异的并行可扩展性,同时能够容纳通用边界条件。然而,FE基DFT中的主要计算瓶颈来自于迭代求解器中对试函数向量块的离散稀疏哈密顿量的反复应用。传统稀疏矩阵-向量乘法和FE单元-矩阵方法在更高阶多项式中遇到内存限制和高数据移动开销,通常用于DFT计算。为克服这些挑战,本工作开发了用于FE离散化DFT的无矩阵算法,通过在飞地操作中利用结构化张量收缩来加速这些乘积,利用1D基函数和四阶数据。引入了一种统一的多级批量数据布局,能够处理实值和复值算子,以最大化Frontier(AVX2)、Param Pravega(AVX512)和Fugaku(SVE)上的缓存重复使用和SIMD利用。我们还结合了优化缓存重复使用的项,偶奇分解以减少FLOP,以及混合精度内联。广泛的基准测试显示,对于大型多向量伪势DFT计算,无矩阵内核在状态最先进的单元矩阵方法基线上的速度提升为1.5-4倍。对于所有电子DFT计算,无矩阵算子由于其高效的实现和优越的算术强度,可获得高达5.8倍的提升。当与容错的Chebyshev过滤子空间迭代求解器结合时,无矩阵形式在FE网格中实现了所需的基态性质的精度,从而显著减少了端到端的求解时间。

英文摘要:

Finite-element (FE) discretisations have emerged as a powerful real-space alternative to large-scale Kohn-Sham density functional theory (DFT) calculations, offering systematic convergence, excellent parallel scalability, while accommodating generic boundary conditions. However, the dominant computational bottleneck in FE-based DFT arises from the repeated application of the discretised sparse Hamiltonian to large blocks of trial vectors during iterations in an iterative eigensolver. Traditional sparse matrix-vector multiplications and FE cell-matrix approaches encounter memory limitations and high data-movement overheads, particularly at higher polynomial orders, typically used in DFT calculations. To overcome these challenges, this work develops matrix-free algorithms for FE-discretised DFT that substantially accelerate these products by doing on-the-fly operations that utilize structured tensor contractions over 1D basis functions and quadrature data. A unified multilevel batched data layout that handles both real and complex-valued operators is introduced to maximise cache reuse and SIMD utilisation on Frontier (AVX2), Param Pravega (AVX512) and Fugaku (SVE). We also combine terms for optimal cache reuse, even-odd decomposition to reduce FLOP, and mixed-precision intrinsics. Extensive benchmarks show that for large multivector pseudopotential DFT calculations, the matrix-free kernels deliver 1.5-4x speedups over the state-of-the-art cell-matrix approach baselines. For all-electron DFT calculations, the matrix-free operator achieves gains of up to 5.8x due to its efficient implementation and superior arithmetic intensity. When integrated with an error-tolerant Chebyshev-filtered subspace iteration eigensolver, the matrix-free formalism yields substantial reductions in end-to-end time-to-solution using FE meshes that deliver desired accuracies in ground-state properties.

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。

↑