arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1702.08880cs.CEcs.MS
收藏

新兴架构上采用自适应网格细化的 Landau 碰撞积分求解器

Landau Collision Integral Solver with Adaptive Mesh Refinement on Emerging Architectures

M. F. Adams, E. Hirvijoki, M. G. Knepley, J. Brown, T. Isaac, R. Mills

AI总结:

本文针对聚变等离子体中的多物种 Landau 碰撞积分,提出结合 PETSc 与自适应网格细化的守恒高阶有限元求解方法,并通过 AVX-512 向量化在 Knights Landing 上实现高效计算。

摘要:

Landau 碰撞积分是聚变等离子体中以小角度散射为主的 Coulomb 碰撞的精确模型。我们使用 PETSc 库(www.mcs.anl.gov/petsc),研究非线性多物种 Landau 积分的一种高精度、完全守恒的有限元离散,并采用自适应网格细化。我们开发了高效利用新兴架构的算法和技术,其方法可最小化内存使用和数据移动,并适合向量处理。Landau 碰撞积分通过 Intel AVX-512 内联函数进行向量化,该求解器在第二代 Intel Xeon Phi 处理器 Knights Landing 上可持续达到高达理论峰值浮点运算率 22% 的性能。

英文摘要:

The Landau collision integral is an accurate model for the small-angle dominated Coulomb collisions in fusion plasmas. We investigate a high order accurate, fully conservative, finite element discretization of the nonlinear multi-species Landau integral with adaptive mesh refinement using the PETSc library (www.mcs.anl.gov/petsc). We develop algorithms and techniques to efficiently utilize emerging architectures with an approach that minimizes memory usage and movement and is suitable for vector processing. The Landau collision integral is vectorized with Intel AVX-512 intrinsics and the solver sustains as much as 22% of the theoretical peak flop rate of the Second Generation Intel Xeon Phi, Knights Landing, processor.

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。

↑