A Effective Algorithm of Hardware Synchronization on Network Processor

被引:0
|
作者
Li Kang [1 ]
Ma Pei-Jun [1 ]
Shi Jiang-Yi [1 ]
机构
[1] Xidian Univ, Def Key Lab Wide Band Gap Semicond Mat & Devices, Xian 710071, Peoples R China
关键词
Computer architecture; Network processor; Hardware Synchronization; Multithreading; MULTIPROCESSORS;
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
The performance of multithreaded network processor can be lowered when its overhead of synchronization increase. So conventional "busy-wait" synchronization algorithm restricted severely the parallel processing capability. A blocking synchronization algorithm of multithreading was proposed for the Multi-Processor System on Chip(MPSoC), which was used to actively schedule exclusive access by the synchronous arbiter on order of first in first out (FIFO). The algorithm was implemened with hardware structure which consisted of the content addressing based lock unit and synchronous request buffer. The implementation of the algorithm is able to eliminate the overhead of memory access in "busy-wait" synchronization and reduce occupancy of on-chip bus bandwidth and processores resource. Comparing with the other algorithm used in the application of packet processing, the algorithm proposed improved about 54% of the performance of the target system with reducing 29% of system power dissipation, which proved effective improvement for performance of packet processing and power dissipation.
引用
收藏
页码:102 / 106
页数:5
相关论文
共 13 条
  • [1] Chen Guo-Liang, 2008, Chinese Journal of Computers, V31, P1493, DOI 10.3724/SP.J.1016.2008.01493
  • [2] [刘亮 LIU Liang], 2008, [通信学报, Journal on Communications], V29, P40
  • [3] Nepsim: A network processor simulator with a power evaluation framework
    Luo, Y
    Yang, J
    Bhuyan, LN
    Zhao, L
    [J]. IEEE MICRO, 2004, 24 (05) : 34 - 44
  • [4] ALGORITHMS FOR SCALABLE SYNCHRONIZATION ON SHARED-MEMORY MULTIPROCESSORS
    MELLORCRUMMEY, JM
    SCOTT, ML
    [J]. ACM TRANSACTIONS ON COMPUTER SYSTEMS, 1991, 9 (01): : 21 - 65
  • [5] MEMIK G, 2001, NETBENCH BENCHMARKIN
  • [6] Efficient synchronization for embedded on-chip multiprocessors
    Monchiero, Matteo
    Palermo, Gianluca
    Silvano, Cristina
    Villa, Oreste
    [J]. IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS, 2006, 14 (10) : 1049 - 1062
  • [7] MUDIGONDA J, 2005, OVERCOMING MEMORY WA
  • [8] PATTERSON DA, 2003, COMPUTER ARCHITECTUR, P584
  • [9] QIN W, 2003, FLEXIBLE FORMAL MODE
  • [10] VILLA O, 2005, FAST DYNAMIC MEMORY