Comparison of Preconditioning Strategies in Energy Conserving Implicit Particle in Cell Methods

Lorenzo Siddi, E. Cazzola, G. Lapenta
{"title":"Comparison of Preconditioning Strategies in Energy Conserving Implicit Particle in Cell Methods","authors":"Lorenzo Siddi, E. Cazzola, G. Lapenta","doi":"10.4208/cicp.oa-2017-0171","DOIUrl":null,"url":null,"abstract":"This work presents a set of preconditioning strategies able to significantly accelerate the performance of fully implicit energy-conserving Particle-in-Cell methods to a level that becomes competitive with semi-implicit methods. We compare three different preconditioners. We consider three methods and compare them with a straight unpreconditioned Jacobian Free Newton Krylov (JFNK) implementation. The first two focus, respectively, on improving the handling of particles (particle hiding) or fields (field hiding) within the JFNK iteration. The third uses the field hiding preconditioner within a direct Newton iteration where a Schwarz-decomposed Jacobian is computed analytically. Clearly, field hiding used with JFNK or with the direct Newton-Schwarz (DNS) method outperforms all method. We compare these implementations with a recent semi-implicit energy conserving scheme. Fully implicit methods are still lag behind in cost per cycle but not by a large margin when proper preconditioning is used. However, for exact energy conservation, preconditioned fully implicit methods are significantly easier to implement compared with semi-implicit methods and can be extended to fully relativistic physics.","PeriodicalId":8424,"journal":{"name":"arXiv: Computational Physics","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2018-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv: Computational Physics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.4208/cicp.oa-2017-0171","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2

Abstract

This work presents a set of preconditioning strategies able to significantly accelerate the performance of fully implicit energy-conserving Particle-in-Cell methods to a level that becomes competitive with semi-implicit methods. We compare three different preconditioners. We consider three methods and compare them with a straight unpreconditioned Jacobian Free Newton Krylov (JFNK) implementation. The first two focus, respectively, on improving the handling of particles (particle hiding) or fields (field hiding) within the JFNK iteration. The third uses the field hiding preconditioner within a direct Newton iteration where a Schwarz-decomposed Jacobian is computed analytically. Clearly, field hiding used with JFNK or with the direct Newton-Schwarz (DNS) method outperforms all method. We compare these implementations with a recent semi-implicit energy conserving scheme. Fully implicit methods are still lag behind in cost per cycle but not by a large margin when proper preconditioning is used. However, for exact energy conservation, preconditioned fully implicit methods are significantly easier to implement compared with semi-implicit methods and can be extended to fully relativistic physics.
隐式粒子胞法节能预处理策略比较
这项工作提出了一套预处理策略,能够显着加快全隐式节能粒子细胞方法的性能,达到与半隐式方法竞争的水平。我们比较三种不同的预处理器。我们考虑了三种方法,并将它们与直接的无前提雅可比自由牛顿克雷洛夫(JFNK)实现进行了比较。前两个关注点分别是在JFNK迭代中改进粒子(粒子隐藏)或字段(字段隐藏)的处理。第三种方法在直接牛顿迭代中使用域隐藏预条件,其中schwarz分解的雅可比矩阵是解析计算的。显然,与JFNK或直接牛顿-施瓦茨(DNS)方法一起使用的字段隐藏优于所有方法。我们将这些实现与最近的半隐式节能方案进行比较。完全隐式方法在每个周期的成本上仍然落后,但在使用适当的预处理时不会有很大的差距。然而,对于精确的能量守恒,预条件全隐式方法比半隐式方法更容易实现,并且可以推广到完全相对论物理中。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信