期刊文献+

多体问题FMM算法在加速部件FPGA研究与实现 被引量:4

Research and realization of N-body problem on FPGA accelerator based on FMM algorithm
下载PDF
导出
摘要 针对FPGA能较好满足高性能计算的异构多核、并行、低成本、低能耗要求,研究了高性能计算的重要的应用之一——多体问题。分析了多体问题应用广泛的FMM算法以及FMM算法的各个算粒,并在FPGA器件实现算粒,与多核CPU上实现这些算粒进行比较,FPGA都获得了不错的加速比。分析了FPGA应用高性能计算的一些优势和当前面临的问题,对FPGA广泛应用高性能计算进行了初步探索。 FPGA as the accelerator of the high-performance calculation can meet multi-core, parallel processing, low costing and lowpower consumption in high-performance calculation. A important application of high performance calculation-N-body problem is studied, and the fast multipole method (FMM) algorithm is analyzed. Several different parts of FMM algorithm are implemented on FPGA and the performance is compared which implement multi-core CPU and FPGA. And then, FPGA obtains the better perfect performance. At last, the advantages and disadvantages of FPGA are analyzed that apply on high performance calculation.
出处 《计算机工程与设计》 CSCD 北大核心 2011年第10期3391-3394,共4页 Computer Engineering and Design
基金 上海市重点学科建设基金项目(J50103)
关键词 高性能计算 多体问题 FMM(快速多极算法) FPGA(现场可编程门阵列) 协处理器 high performance calculation N-bodyproblem FMM FPGA coprocessor
  • 相关文献

参考文献15

  • 1Bill Dally. Life after moore's law [DB/OL] .http://www.forbes. corn/2010/04/29/moores-law-computing-processing-opinions- contributors-bill-dally_2.html,2010-04-29/2010-07-20.
  • 2Matsuo K,Hamada T, Miyoshi M,et al.Accelerating phase corre- lation functions using GPU and FPGA [C]. Proceedings of the NASA/ESA Conference on Adaptive Hardware and Systems, 2009:433-438.
  • 3Kapre N,DeHon A.Performance comparison of single-precision SPICE Model-Evaluation on FPGA, GPU, Cell, and multi-core processors[C].International Conference on Field Programmable Logic and Applications,2009:65-72.
  • 4Storaasli O,Strenski D.Exploring accelerating science applica- tions with FPGAs[R].The Reconfigurable Systems Summer In- stitute,2007.
  • 5Tarek EI-Ghazawi, Esam El-Araby, Miaoqing Huang, et al. The promise of high-performance reconfigurable computing [J].Computer,2008,41 (2):69-76.
  • 6Greengard L,Rokhlin V.A fast algorithm for particle simulations [J].Journal of Computational Physics, 1987,73(2):325-348.
  • 7赖国明,杨圣云,袁德辉.FMM算法的并行化方法[J].计算机应用与软件,2007,24(7):176-178. 被引量:2
  • 8Wain R,Bush I,Guest M,et al.An overview of FPGAs and FPGA programming JR], Initial Experiences at Daresbury, Computa- tional Science and Engineering Department,CCLRC Daresbury Laboratory, 2006.
  • 9Pico Computing Inc,Impulse Accelerated Technologies Inc.Soft- ware-to-FPGA workstation unveiled[DB/OL].http://www, hpcwire. com/offthewire/Software-to-FPGA Workstation Unveiled. html? page=1,2008-10-16/2010-07-25.
  • 10Pico Computing Inc. FPGA cluster accelerates bioinformatics application by 5000X[DB/OL].http://www.hpcwire.com/offthe- wire/FPGA-Cluster-Accelerates-Bioinformatics-Application- by-5000X-69612762.html,2009-11-09/2010-07-25.

二级参考文献11

  • 1赖国明,杨圣云,袁德辉.关于快速多极算法FMM的几点注解[J].现代计算机,2006,12(3):25-29. 被引量:1
  • 2赖国明,杨圣云.多极扩展与Barnes-Hut算法的误差分析[J].福建电脑,2006,22(5):17-18. 被引量:1
  • 3Greengard L,Rokhlyn V.1987,J.Comp.Phys.73,325.
  • 4Michael S Warren,John K Salmon.A Parallel Hashed Oct-Tree N-Body Algorithm.Supercomputing'93,Califormia Institute of Technology Pasdena,CA.
  • 5Warren M S,Salmon J K.Fast Parallel Tree Codes for Gravitational and Fluid Dynamical N-Body Problems.submitted to the International Journal of Supercomputing,1992,8(2).
  • 6Okunbor D,JuiLinlu E.Parallel Fast Multipole Algorithm using MPI.University of Missouri-Rolla,1995,6.
  • 7Cheng H,Greengard L,Rokhlin V.A fast adaptive multipole algorithm in three dimensions.Journal of Computational Physics,1999:468498.
  • 8William T Rankin III.Efficient Parallel ImpleMentations of Multipole Based N-Body Algorithms.Doctor Thesis,1999.
  • 9Ying Lexing,Biros G,Zorin D.A kernel-independent fast multipole algorithm.Technical Report TR2003-839.New York University,New York 2003,4.
  • 10Sarin V,Grama A,Sameh A.Analyzing the Error Bounds of Multipole-Based Treecodes.Purdue Univ.W.Lafayette,1998,9:1-12.

共引文献1

同被引文献49

  • 1耿涛,刘经南,赵齐乐,施闯.星地监测网下的北斗导航卫星轨道确定[J].测绘学报,2011,40(S1):46-51. 被引量:14
  • 2柳响林,Pavel Ditmar.基于B-spline和正则化算法的低轨卫星轨道平滑[J].地球物理学报,2006,49(1):99-105. 被引量:10
  • 3Prasanna Sundararajan. High Performance Computing Using FPGAs XILINX White Paper[ OL]. WP375,2010.
  • 4Dimond Rob, Racanière Srbastien, Pell Oliver. Accelerating Large- Scale HPC Applications Using FPGAs[ C]//IEEE 2011. Germany : Proceedings - 2011 20th Symposium on Computer Arithmetic ,2011 : 191 - 192.
  • 5罗兴国,等.PRCA:一种高效能计算体系结构[C]//2012高效能计算机体系结构国际高端论坛,上海,2012,10.
  • 6Xilinx. Virtex-5 Family Overview. Xilinx Product Specification DS100 [OL]. http ://www. xilinx. com/2012.
  • 7Xilinx. 7 Series FPGAs Overview. Xilinx Advance Product Specification DS180[OL]. http://www. xilinx. com/2012.
  • 8John Hennessy,David Patterson. Computer Architecture: A Quantita- tive Approach[ M ]. 4th ed. Morgan Kaufmann,2006.
  • 9Zhe Zheng, Yongxin Zhu, Xu Wang, et al. Revealing Feasibility of FMM on ASIC: Efficient Implementation of N-Body Problem on FPGA [ C ]//IEEE International Conference on Computational Science and Engineering, Hang Kong. 2010 : 132 - 139.
  • 10刘林.人造地球卫星轨道力学[M]{H}北京:高等教育出版社,1992.

引证文献4

二级引证文献5

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部