期刊文献+

部分可重构系统布局的一种新算法 被引量:4

A new methodology for placement on partial reconfigurable devices
下载PDF
导出
摘要 针对动态部分可重构系统的瓶颈,即布局算法必须在保证运行速度的基础上,尽可能增加可重构芯片利用率的问题,提出了一种布局算法KVIT(keeping the vertexes information of tasks).其核心思想是尝试将新到达的硬件任务放置在已布局硬件任务的顶点处,并通过对可重构芯片内部计算单元进行编码迅速判断新任务是否可放置在该顶点.该算法的时间复杂度为O(N),N是可重构系统中当前运行的硬件任务的数目.仿真实验结果表明,KVIT算法的布局质量与现有的O(N2)时间复杂度布局算法基本一致,而其执行速度则明显高于已有算法. The fact that placement algorithm has to obtain high total chip utilization and low time complexity simultaneously constitutes a bottleneck in partially reconfigurable systems. A KVIT (keeping the vertexes information of tasks) algorithm with time complexity O(N) is thus presented, where N is the number of hardware tasks currently running on the chip. The main idea is to put newly arrived tasks on the vertexes of the already running tasks, and then quickly validate this placement using an encoding scheme of the reconfigurable hardware. Simulation resultS show that the KVIT algorithm is obviously more efficient, while providing almost the same task reiection ratio compared with the other O(N2) algorithms.
出处 《中国科学技术大学学报》 CAS CSCD 北大核心 2007年第9期1047-1053,共7页 JUSTC
基金 国家自然科学基金(60273042) 中国科学院创新基金 安徽省自然科学基金(03042203)资助
关键词 可重构 布局算法 调度 任务顶点 reconfigurable placement algorithm schedule task vertexes
  • 相关文献

参考文献7

  • 1Bazargan K, Kastner R, Sarrafzadeh M. Fast template placement for reeonfigurable computing systems [J]. IEEE Design and Test of Computers, 2000, 17(1): 68-83.
  • 2Walder H, Steiger C, Platzner M Fast online task placement on FPGAs: free space partitioning and 2-D hashing[C].Proceedings of the 17th International Parallel and Distributed Processing Symposium ( IPDPS )/Reconfigurable Architectures Workshop (RAW). New York: IEEE Computer Society, 2003 : 178.
  • 3Walder H, Steiger C, Platzner M, et al. Online scheduling and placement of real-time tasks to partially reconfigurable devices [ C ].Real-Time Systems Symposium. Now York:IEEE Computer Society, 2003, 224-225.
  • 4Ahmadinia A, Bobda C, J urgen T. A new approach for on-line placement on reconfigurable devices[C].Proceedings of the 18th International Parallel and Distributed Processing Symposium Stanta Fe, New Mexico: IEEE Computer Society, 2004:1340-1346.
  • 5Handa M, Vemuri R. An efficient algorithm for finding empty space for online FPGA placement[C].Design Automation Conference. New York:ACM Press, 2004: 960-965.
  • 6方潜生,王煦法,何劲松.硬件进化的快速算法模型研究[J].中国科学技术大学学报,2003,33(5):612-618. 被引量:5
  • 7周博,王石记,邱卫东,彭澄廉.SHUM-UCOS:基于统一多任务模型可重构系统的实时操作系统[J].计算机学报,2006,29(2):208-218. 被引量:32

二级参考文献24

  • 1Smith S F, A Learning System Based on Genetic Adaptive Algorithms [ D ]. Doctoral dissertation, University of Alabama , Tuscaloosa,1980.
  • 2Holland J H and Reitman J S. Cognitive System Based on Adaptive Algorithms, Pattern Directed Inference Systems [ M ]. ( Ede. D A Waterman and F Hayes Roth) New York: Academic Press. 1987,313 : 329.
  • 3Garis H D. Evolvable Hardware : Genetic Programming d a Dawin Machine [ A ]. In Proceeding of Artificial Neural Nets and Genetic Algorithms [ C ]. Innsbruck, Austria: Springer Verlag, 1993.
  • 4Yao Xin, Higuchi T. Promises and challenges of evolvable hardware [ J ]. IEEE Trans. On Systems. Man and Cybemetics-PartC : Applications and Reviews. 1999,29( 1 ) :87-97.
  • 5Kajitani I, et al. Variable Length Chromosome GA for Evolvable Hardware[ A ]. In Proc. of 3rd Int. Conf. on Evolutionary Computation [ C ]( ICEC96 ). IEEE Press, Piscataway, NJ,USA, 1996: 443-447.
  • 6Higuchi T, Evolvable Hardware for Industrial Applications [ C ]. The Third NASA/ DoD Workshop on Evolvable Hardware. Long Beach, California, IEEE Press,2001.12-14.
  • 7Lee E..Overview of the Ptolemy Project.Technical Memorandum UCB/ERL M03/25,University of California,Berkeley,CA,USA,2003.
  • 8Alexander P.,Kong C..Rosetta:Semantic support for model centered systems level design.Computer,2001,34(11):64~70.
  • 9Andrews D.,Niehaus D..Programming models for hybrid FPGA-CPU computational components:A missing link.IEEE Transactions on Micro,2004,24(4):42~53.
  • 10Walder H.,Platzner M..Reconfigurable hardware operating systems:From design concepts to realizations.In:Proceedings of the 3rd International Conference on Engineering of Reconfigurable Systems and Architectures (ERSA'03),Las Vegas(NV),USA,2003.

共引文献35

同被引文献20

  • 1盛蓝平,林涛.采用启发式分支定界的软硬件划分[J].计算机辅助设计与图形学学报,2005,17(3):414-417. 被引量:6
  • 2Steiger C, Walder H, Platzer M. Operating systems for reconfigurable embedded platforms: Online scheduling of real-time tasks[J].IEEE Trans Computer,2004,53(11) : 1393-1407.
  • 3Tsbero J, Septien J, Meeha H, et al. Task Placement Heuristic Based on 3D-Adjacency and Look-ahead in Reconfigurable Systems[C]//IEEE. 2006 : 396-401.
  • 4Redaelli F, Santambrogio M D, Rana V, et al. Scheduling and 2D Placement Heuristics for Partially Recorffigurable Systems[C]//IEEE. 2009 : 223-230.
  • 5Hagemeyer J, Kettelhoit B, Koester M, et al. Design Homogeneous Communication Infrastructures for Partially Reconfigurable FPGAs[A] // Proceedings of the International Conference on Reconfigurable Systems and Algorithms [C]. Las Vegas, 2007 : 238-247.
  • 6Sharma A, Hauck S, Ebeling C. Architecture-adaptive Routability-driven Placement for Fpgas[C]//IEEE. 2005:0-7803-9362-7.
  • 7齐骥,李曦,于海晨,胡楠,龚育昌,王立刚.一种面向动态可重构计算的调度算法[J].计算机研究与发展,2007,44(8):1439-1447. 被引量:15
  • 8周学功,梁樑,黄勋章,彭澄廉.可重构系统中的实时任务在线调度与放置算法[J].计算机学报,2007,30(11):1901-1909. 被引量:27
  • 9Bazargan K,Kastner R,Sarrafzadeh M. Fast Template Placement For Reconfigurable Computing Systems[J].IEEE Design and Test of Computers,2000,(01):68-83.
  • 10Walder H,Steiger C,Platzner M. Fast Online Task Placement on FPGAs:Free Space Partitioning and 2D hashing[A].Washington D C,USA:IEEE Computer Society,2003.178.

引证文献4

二级引证文献9

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