Platform and algorithm effects on computational fluid dynamics applications in life sciences
详细信息 查看全文
- 作者：Sofia Guzzetti ; sofia.guzzetti@emory.edu" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Tiziano Passerini1 ; tpasser@emory.edu" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Jaroslaw Slawinski2 ; jslawin@emory.edu" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Umberto Villa3 ; uvilla@emory.edu" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Alessandro Veneziani avenez2@emory.edu" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Vaidy Sunderam vss@emory.edu" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae
- 关键词：Applied computing in medicine ; Platform heterogeneity ; Cloud computing ; Computational fluid dynamics ; Parallel overlapping domain decomposition ; Performance analysis
- 刊名：Future Generation Computer Systems
- 出版年：2017
- 出版时间：February 2017
- 年：2017
- 卷：67
- 期：Complete
- 页码：382-396
- 全文大小：2522 K
文摘
- •
Methodologies for the optimal choice of computing platforms for CFD are proposed.
- •
The optimal partitioning of 3D domains with overlapping DD is discussed.
- •
IaaS resources may be utilized for scientific CFD at a lower cost than incurred locally.
- •
MPI-based performances can be adapted by communication-aware placement of processes.
- •
A thoughtful selection of small overlap can benefit the overall computational time.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |