Single processor optimization of OVERFLOW
暂无分享,去创建一个
Linux PC Clusters are a cost effective platform for parallel computational fluid dynamics (CFD) applications. In this paper a variety of node constructions are considered, from IA32 commodity processors to 64-bit MP (Multi-Processor) and mutli-core configurations. Single processor performance is evaluated and the most likely subroutines for optimization of single processor performance are identified based on cache performance. This information may in turn be used to improve the single processor performance of OVERFLOW.
[1] William I. Scallion,et al. CFD Approaches for Simulation of Wing-Body Stage Separation , 2004 .
[2] Raymond P. LeBeau,et al. Use of Cache-Friendly Blocking to Accelerate CFD Codes on Commodity Hardware , 2006 .
[3] Raymond P. LeBeau,et al. Improving the Performance of Computational Fluid Dynamics Codes on Linux Cluster Architectures , 2003 .