| 基于混合架构的FMM算法硬件加速 |
| |
| 引用本文： | 曹旻,李海强,曹真.基于混合架构的FMM算法硬件加速[J].计算机工程,2012,38(16):275-278. |
| |
| 作者姓名： | 曹旻 李海强 曹真 |
| |
| 作者单位： | 上海大学计算机工程与科学学院 |
| |
| 基金项目： | 国家“863”计划基金资助项目(2009AA012201-CFA2009SHDX01);上海市重点学科建设基金资助项目(J50103) |
| |
| 摘 要： | 以高性能计算中的经典问题——多体问题的快速多极子(FMM)算法为例，分析FMM算法的各个步骤，根据计算、通信和存储特性将算法中的子过程归类。在CPU、GPU、FPGA和CELL上分别进行测试，提出执行FMM算法的混合可重构体系结构配置方案，并进一步优化算法，分解任务流。针对不同任务流的特点，提出可行的解决方案。结果证明，该方案可提高算法效率。
|
| 关 键 词： | 混合可重构计算机体系结 加速部件 N-Body问题 快速多极子算法 配置方案 任务流 |
| 收稿时间： | 2011-08-30 |
| 修稿时间： | 2011-12-06 |
Hardware Acceleration of FMM Algorithm Based on Mixed Architecture |
| |
| CAO Min,LI Hai-qiang,CAO Zhen.Hardware Acceleration of FMM Algorithm Based on Mixed Architecture[J].Computer Engineering,2012,38(16):275-278. |
| |
| Authors: | CAO Min LI Hai-qiang CAO Zhen |
| |
| Affiliation: | (School of Computer Engineering and Science,Shanghai University,Shanghai 200072,China) |
| |
| Abstract: | Accelerators are increasingly viewed as computer coprocessors that can provide significant computational performance at low price.This paper implements and tests every sub-procedure of Fast Multipole Method(FMM) on GPU,FPGA and CELL based on the analysis of computational,storage and communication characteristics.It makes two contributions to optimize FMM.A mixed configurable computer architecture which can run FMM well is presented.FMM is optimized on mixed architecture through decomposing its task flow.The probable solution for different task flow is also put forward based on the large experiment results.Results show that the scheme can increase the efficiency of the algorithm. |
| |
| Keywords: | mixed configurable computer architecture acceleration component N-Body problem Fast Multipole Method(FMM) algorithm configuration scheme task flow |
| 本文献已被 CNKI 维普 等数据库收录！ |
| 点击此处可从《计算机工程》浏览原始摘要信息 |
|
点击此处可从《计算机工程》下载全文 |
