Redefining the Relationship between Scalar and Parallel Units in SIMD Architectures

被引:0
|
作者
Wang, Yaohua [1 ]
Chen, Shuming [1 ]
Wan, Jianghua [1 ]
Zhang, Kai [1 ]
机构
[1] Natl Univ Def Technol, Sch Comp Sci, Changsha, Hunan, Peoples R China
关键词
SIMD; loosely coupled; tightly coupled;
D O I
暂无
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
SIMD architectures, comprising of both scalar and parallel units, have been widely used in media processors. To further improve the performance, much effort has been made to enhance the design of both units, while little attention has been placed on the relationship between the units. This paper demonstrates that a dynamic coupling mechanism, which can dynamically transform the scalar and parallel units between loosely and tightly coupled relationships, can achieve a better matching between existing SIMD architectures and media applications. The evaluation result shows that the dynamic coupling mechanism can achieve an average performance gain of 33.2% at an additional area cost of 4.57%.
引用
收藏
页码:781 / 784
页数:4
相关论文
共 50 条
  • [1] A parallel join algorithm for SIMD architectures
    Azadegan, S
    Tripathi, A
    JOURNAL OF SYSTEMS AND SOFTWARE, 1997, 39 (03) : 265 - 280
  • [2] Scalar Processing Overhead on SIMD-Only Architectures
    Azevedo, Arnaldo
    Juurlink, Ben
    2009 20TH IEEE INTERNATIONAL CONFERENCE ON APPLICATION-SPECIFIC SYSTEMS, ARCHITECTURES AND PROCESSORS, 2009, : 183 - 190
  • [4] A PARALLEL INERTIA METHOD FOR FINDING EIGENVALUES ON VECTOR AND SIMD ARCHITECTURES
    CONROY, JM
    PODRAZIK, LJ
    SIAM JOURNAL ON SCIENTIFIC COMPUTING, 1995, 16 (02): : 500 - 505
  • [5] SIMD Defragmenter: Efficient ILP Realization on Data-parallel Architectures
    Park, Yongjun
    Seo, Sangwon
    Park, Hyunchul
    Cho, Hyoun Kyu
    Mahlke, Scott
    ASPLOS XVII: SEVENTEENTH INTERNATIONAL CONFERENCE ON ARCHITECTURAL SUPPORT FOR PROGRAMMING LANGUAGES AND OPERATING SYSTEMS, 2012, : 363 - 374
  • [6] SIMD Defragmenter: Efficient ILP Realization on Data-parallel Architectures
    Park, Yongjun
    Seo, Sangwon
    Park, Hyunchul
    Cho, Hyoun Kyu
    Mahlke, Scott
    ACM SIGPLAN NOTICES, 2012, 47 (04) : 363 - 374
  • [7] Dual-Core Framework: Eliminating the Bottleneck Effect of Scalar Kernels on SIMD Architectures
    Wang, Yaohua
    Chen, Shuming
    Chen, Hu
    Wan, Jianghua
    Zhang, Kai
    Liu, Sheng
    IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2013, E96D (02): : 365 - 369
  • [8] PARALLEL CODE GENERATION FOR SUPER-SCALAR ARCHITECTURES
    FERNANDES, EST
    BORNSTEIN, CF
    PEREIRA, CMD
    MICROPROCESSING AND MICROPROGRAMMING, 1992, 34 (1-5): : 223 - 226
  • [9] Real-time filtering on parallel SIMD architectures for automated quality inspection
    Usamentiaga, R.
    JOURNAL OF REAL-TIME IMAGE PROCESSING, 2021, 18 (01) : 127 - 141
  • [10] Structured sparse matrix-vector multiplication on massively parallel SIMD architectures
    Dehn, T
    Eiermann, M
    Giebermann, K
    Sperling, V
    PARALLEL COMPUTING, 1995, 21 (12) : 1867 - 1894