期刊文献+

一种共享主存二维SIMD结构资源分配算法的改进与实现

Improvement and Implementation of a Resource Allocation Algorithm for the Shared Memory 2D SIMD Architecture
下载PDF
导出
摘要 共享主存二维SIMD结构已经广泛应用于多媒体处理加速部件,其数据并行性可以大大提高处理器的运算能力。目前,已有一些针对共享主存二维SIMD结构编译优化方面的研究,这些编译优化技术能有效地提高各种多媒体应用程序的加速比。但是,分析可知,这些优化方法的平均资源利用率只有约50%。本文基于对多媒体应用程序在共享主存二维SIMD结构上的执行过程分析,根据原有算法并适当修改经典图着色寄存器分配算法,提出了一种改进的资源分配方法,以提高SIMD单元的平均利用率,最终达到提高加速比的目的。实验结果说明,该算法的改进对于大部分多媒体应用程序的性能有显著的提高。 The shared-memory 2D SIMD architecture is widely used as a multimedia accelerator in multimedia processing fields, and it can improve the computing ability of processors for its data parallelisrm There are some research on the optimization techniques which can improve the speedup of multimedia applications for the shared memory 2D SIMD architecture. However, their average resource utilization rate is less than 50%. In this paper, we first analyze the characteristics of the applications' execution process on the shared memory 2D SIMD architecture, and then propose an improved resource allocation method based on the original one in purpose of improving the average utilization rate of the SIMD units. Our experimental results show that our improved algorithm is useful for the performance improvement of most multimedia applications.
出处 《计算机工程与科学》 CSCD 2008年第9期99-102,136,共5页 Computer Engineering & Science
关键词 二维SIMD 共享主存 资源分配 2D SIMD shared memory resource allocation
  • 相关文献

参考文献9

  • 1Singh H, Lee M H, Lu G, et al. MorhpoSys: An Integrated Reconfigurable System for Data-Parallel and Computation Intensive Applications[J]. IEEE Trans on Computers, 2000, 49(5) :465-481.
  • 2Komuro T, Kagami S, Ishikawa M. A Dynamically Reconfigurable SIMD Processor for a Vision Chip[J]. IEEE Journal of Solid-State Circuites, 2004, 39(1): 265-268.
  • 3Miyamori T,Olukotun K. A Quantitative Analysis of Reconfigurable Coprocessors for Multimedia Applications [C]// Proc of IEEE Syrup on FPGAs for Custom Computing Machines, 1998.
  • 4Zhang W, Qian X, Wang Y, et al. Optimizing Compiler for Shared Memory Multiple SIMD Architecture[C]///Proe of ACM SIGPLAN/SIGBED Conf on Languages, Compilers, and Tools for Embedded Systems, 2006.
  • 5Zhang W, Bao T, Zang B, et al. Data Pipeline Optimization for Shared Memory Multiple SIMD Arehitecture[C]//Proe of the 19th Int'l Workshop on Languages and Compilers for Parallel Computing, 2006.
  • 6张为华,臧斌宇,王晔,钱兴隆,朱传琪.基于共享向量的二维SIMD调度算法[J].计算机学报,2006,29(10):1740-1749. 被引量:2
  • 7张为华,王晔,鲁云萍,臧斌宇,朱传琪.一种可重配置系统的模型[J].小型微型计算机系统,2006,27(7):1245-1249. 被引量:7
  • 8Hofstee H P. Power Efficient Processor Architecture and the Cell Processor[C]//Proc of the 11th Int'l Conf on High-Performance Computer Architecture,2005.
  • 9Talla D, John L K, Burger D. Bottlenecks in Multimedia Processing with SIMD Style Extensions and Architectural Enhancements[J]. IEEE Trans on Computers, 2003, 52(8) : 1015-1031.

二级参考文献27

  • 1Milagros Fernández,Nader Bagherzadeh,Rafael Maestre,Roman Hermida,Fadi Kurdahi,Marcos Sanchez-Elez.A data scheduler for multi-context reconfigurable architectures[C].Proceedings of the 14th international symposium on Systems synthesis,Montreal,P.Q.,Canada,September 2001,177-182.
  • 2Maestre R,Fernandez M,Hermida R,Bagherzadeh N.A framework for scheduling and context allocation in reconfigurable computing[C].12th International Symposium on System Synthesis,Boca Raton,Florida,November 1999.
  • 3Chengzhi Pan,Nader Bagherzadeh,Amir Hosein Kamalizad,Arezou Koohi.Design and analysis of a programmable single-chip architecture for DVB-T base-band receiver[C].Design,Automation and Test in Europe.Conference and Exhibition (DATE'03),Munich,Germany,2003:468-472.
  • 4Intel.IntelrPentiumr 4 and Intelr XeonTM processor optimization reference manual[EB/OL].http://www.intel.com/cd/ids/ developer/asmo-na/eng/microprocessors/ia32/pentium4/index.htm 2004.
  • 5Chen W H,Smith C H,Fralick S C.A fast computational algorithm for the discrete cosine transform[J].IEEE Transaction on Communications,1977,25(9):1004-1009.
  • 6Singh H,Lee M H,Lu G,Kurdahi F J,Bagherzadeh N,Filho E M C.MorphoSys:an integrated reconfigurable system for data-parallel and computation -intensive applications[J].IEEE Transaction on Computers,2000,49(5):465-481.
  • 7Lu G,Singh H,Lee M,et al.The morphoSys dynamically reconfigurable system-on-chip[C].In:The First NASA/DOD Workshop on Evolvable Hardware,Pasadena,California,1999,152-161.
  • 8Singh H,Ming-Hau Lee,Guangming Lu,Kurdahi F J,Bagherzadeh N,Lang T,Heaton R,Filho E M C.MorphoSys:an Integrated Re-Configurable Architecture.The application of information technologies (Computer Science) to mission systems[M].Monterey,CA,USA,April 1998.
  • 9Lu Guang-ming,Ming-Hau Lee,Nader Bagherzadeh,Rafael Maestre,Eliseu Filho,Fadi Kurdahi,Hartej Singh MorphoSys:case study of a reconfigurable computing system targeting multimedia applications[C].37th Conference on Design Automation (DAC'00) June 2000,Los Angeles,CA,573-578.
  • 10Maestre R,Kurdahi F J,Bagherzadeh N,Singh H,Hermida R,Fernandez M.Kernel scheduling in reconfigurable computing[C].In:Design,Automation and Test in Europe (DATE '99) Munich,Germany,1999,90-97.

共引文献7

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