Multilevel Space-Time Additive Schwarz Methods for Parabolic Equations

Multilevel Space-Time Additive Schwarz Methods for Parabolic Equations
复制标题

抛物线方程的多级时空加性SCHWARZ方法

DOI:
10.1137/17m113808x
复制
发表时间:
2018-09
影响因子:
3.1
通讯作者:
Cai Xiao Chuan
Cai Xiao Chuan
中科院分区:
数学2区
文献类型:
--
作者:
Li Shishun;Shao Xinping;Cai Xiao Chuan

文献摘要

参考文献

被引文献

相似文献

本文提出了一种多层时空加性施瓦茨方法,用于求解由抛物型方程离散化所产生的线性方程组。利用这种方法,问题在空间和时间维度上并行求解。在建立了空间和时间分解的两个重要性质之后,即,一个加强的Cauchy-Schwarz型不等式和一个稳定的多级分解下的时空能量范数,我们发展了一个最优收敛理论在$R^2$和$R^3$,并显示如何收敛速度依赖于网格大小,子域的数量,窗口大小,和水平的数量。在一台具有数千个处理器的并行计算机上进行的二维和三维问题的数值实验证实了理论的迭代次数,以及强和弱的可伸缩性。与传统的仅在空间上并行的时间步进方法相比,时空方法具有更好的性能。
In this paper, we present a multilevel space-time additive Schwarz method for solving linear system of equations arising from the discretization of parabolic equations. With this method, the problem is solved in parallel on both space and time dimensions. After establishing two important properties of the space and time decomposition, i.e., a strengthened Cauchy--Schwarz-type inequality and a stable multilevel decomposition under a space-time energy norm, we develop an optimal convergence theory in $R^2$ and $R^3$ and show how the convergence rate depends on the mesh sizes, the number of subdomains, the window size, and the number of levels. Numerical experiments carried out on a parallel computer with thousands of processors for two- and three-dimensional problems confirm the theory in terms of the number of iterations, as well as the strong and weak scalabilities. Furthermore, a detailed comparison shows that the space-time method outperforms the traditional time stepping method, parallelized only in sp...
DOI: 10.1109/tcad.1982.1270004
发表时间: 1982-07
影响因子: 2.9
作者:
E. Lelarasmee;A. Ruehli;A. Sangiovanni-Vincentelli
通讯作者: E. Lelarasmee;A. Ruehli;A. Sangiovanni-Vincentelli
DOI: 10.1051/m2an/2013121
发表时间: 2010-06
期刊: arXiv: Analysis of PDEs
影响因子: --
作者:
Minh-Binh Tran
通讯作者: Minh-Binh Tran
DOI: 10.1137/s1064827596305337
发表时间: 1998-11
期刊: SIAM J. Sci. Comput.
影响因子: --
作者:
M. Gander;A. Stuart
通讯作者: M. Gander;A. Stuart
DOI: 10.1007/bf01385713
发表时间: 1991-12
影响因子: 2.1
作者:
X. Cai
通讯作者: X. Cai
DOI: 10.1016/s0096-3003(02)00085-1
发表时间: 2003-03
期刊: Appl. Math. Comput.
影响因子: --
作者:
H. Rui
通讯作者: H. Rui