Delayed approximate matrix assembly in multigrid with dynamic precisions

Delayed approximate matrix assembly in multigrid with dynamic precisions
复制标题

具有动态精度的多重网格中的延迟近似矩阵组装

DOI:
10.1002/cpe.5941
复制
发表时间:
2020
期刊:
Practice and Experience
影响因子:
--
通讯作者:
Murray C
Murray C
中科院分区:
--
文献类型:
--
作者:
Murray C

文献摘要

参考文献

被引文献

相似文献

在任何求解网格上偏微分方程的代码中,系统矩阵的精确装配都是一个重要步骤。我们要么明确地建立一个矩阵,要么在一个没有矩阵的环境中工作,在这种环境中,我们必须能够根据需要快速返回矩阵项。无论哪种方式,由于进入方程的重要材料参数,需要相互依赖的矩阵级联的多网格编码,或者需要在整个求解过程中重新计算矩阵条目或整个方程系统的动态自适应网格细化,因此构建成本可能会很高。我们建议这些构造可以与多网格循环同时进行。初始几何矩阵和低精度集成启动了多网格迭代,而改进的装配数据则在可用时提供给求解器。解决问题的时间得到了改善,因为我们消除了传统上延迟实际计算的昂贵的准备阶段。我们消除了算法延迟。此外,我们从解决方案过程中去同步组装。这种并发级别的无规律增加提高了可伸缩性。众所周知,汇编程序对内存和带宽的要求很高。当我们迭代地改进算子精度时,我们最后建议使用分层有损压缩方案,以便在系统矩阵条目携带很少信息或尚未具有高精度的情况下大幅降低内存占用。
The accurate assembly of the system matrix is an important step in any code that solves partial differential equations on a mesh. We either explicitly set up a matrix, or we work in a matrix‐free environment where we have to be able to quickly return matrix entries upon demand. Either way, the construction can become costly due to nontrivial material parameters entering the equations, multigrid codes requiring cascades of matrices that depend upon each other, or dynamic adaptive mesh refinement that necessitates the recomputation of matrix entries or the whole equation system throughout the solve. We propose that these constructions can be performed concurrently with the multigrid cycles. Initial geometric matrices and low accuracy integrations kickstart the multigrid iterations, while improved assembly data is fed to the solver as and when it becomes available. The time to solution is improved as we eliminate an expensive preparation phase traditionally delaying the actual computation. We eliminate algorithmic latency. Furthermore, we desynchronize the assembly from the solution process. This anarchic increase in the concurrency level improves the scalability. Assembly routines are notoriously memory‐ and bandwidth‐demanding. As we work with iteratively improving operator accuracies, we finally propose the use of a hierarchical, lossy compression scheme such that the memory footprint is brought down aggressively where the system matrix entries carry little information or are not yet available with high accuracy.
在多处理器上求解偏微分方程的异步多级自适应方法:性能结果
DOI: --
发表时间: 1989
期刊: Parallel Computing
影响因子: 1.4
作者:
S. McCormick;D. Quinlan
通讯作者: D. Quinlan
非对称黑盒多重网格,粗化三倍
DOI: --
发表时间: 2012
影响因子: 4.3
作者:
I. Yavneh;Marion Weinzierl
通讯作者: Marion Weinzierl
DOI: --
发表时间: 2010
影响因子: 4.3
作者:
J. Dendy;J. Moulton
通讯作者: J. Moulton
DOI: --
发表时间: 1990
期刊: Attribute Grammars and their Applications
影响因子: --
作者:
D. Knuth
通讯作者: D. Knuth
在多处理器上求解偏微分方程的异步多级自适应方法:基本思想
DOI: --
发表时间: 1989
期刊: Parallel Computing
影响因子: 1.4
作者:
L. Hart;S. McCormick
通讯作者: S. McCormick