Barrier Synchronization for CELL Multi-Processor Architecture

被引：2
|
作者：
Bai, Shuwei [1 ]
Zhou, Qingguo [1 ]
Zhou, Rui [1 ]
Li, Lian [1 ]
机构：
[1] Lanzhou Univ, SISE, Distributed & Embedded Syst Lab, Lanzhou 730000, Peoples R China
来源：
2008 FIRST IEEE INTERNATIONAL CONFERENCE ON UBI-MEDIA COMPUTING AND WORKSHOPS, PROCEEDINGS | 2008年
关键词：
CELL; barrier synchronization; mailbox; signal notification register;
D O I：
10.1109/UMEDIA.2008.4570882
中图分类号：
TP3 [计算技术、计算机技术];
学科分类号：
0812 ;
摘要：
Cell microprocessor is a new multi-processors system, which has been used for sonsumer electronics, multimedia decoding/encoding, compress or uncompressing area[3]. One important development technology for multi-processor system is barrier synchronization, which can improve the system performance if the appreciated barrier is adopted Based three different communication mechanisms (mailbox, dma, and signal notification register) offered by CELL system, we implement three kinds of barrier implementation tools on the CELL multi-processor system. In the paper, we will show the implementation of barrier synchronization tools and compare the barriers from three aspects: spu size, performance, and synchronization information capacity.
引用
收藏
页码：155 / 158
页数：4
相关论文
共 50 条
  • [31] Shared multi-processor scheduling
    Dereniowski, Dariusz
    Kubiak, Wieslaw
    EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 2017, 261 (02) : 503 - 514
  • [32] Multi-Processor Debug in SoC and Processor designs
    Penner, Bill
    2013 INTERNATIONAL SYMPOSIUM ON VLSI DESIGN, AUTOMATION, AND TEST (VLSI-DAT), 2013,
  • [33] Chip multi-processor generator
    Solomatnikov, Alex
    Firoozshahian, Amin
    Qadeer, Wajahat
    Shacham, Ofer
    Kelley, Kyle
    Asgar, Zain
    Wachs, Megan
    Hameed, Rehan
    Horowitz, Mark
    2007 44TH ACM/IEEE DESIGN AUTOMATION CONFERENCE, VOLS 1 AND 2, 2007, : 262 - +
  • [34] Predicting inter-thread cache contention on a chip multi-processor architecture
    Chandra, D
    Guo, F
    Kim, S
    Solihin, Y
    11TH INTERNATIONAL SYMPOSIUM ON HIGH-PERFORMANCE COMPUTER ARCHITECTURE, PROCEEDINGS, 2005, : 340 - 351
  • [35] A reconfigurable network-on-chip architecture for optimal multi-processor SoC communication
    Rana V.
    Atienza D.
    Santambrogio M.D.
    Sciuto D.
    De Micheli G.
    IFIP Advances in Information and Communication Technology, 2010, 313 : 232 - 250
  • [36] Evolvable multi-processor: a novel MPSoC architecture with evolvable task decomposition and scheduling
    Vakili, S.
    Fakhraie, S. M.
    Mohammadi, S.
    IET COMPUTERS AND DIGITAL TECHNIQUES, 2010, 4 (02): : 143 - 156
  • [37] A new architecture for concurrent lazy cyclic reference counting on multi-processor systems
    Formiga, Andrei de Araujo
    Lins, Rafael Dueire
    JOURNAL OF UNIVERSAL COMPUTER SCIENCE, 2007, 13 (06) : 817 - 829
  • [38] A Reconfigurable Network-on-Chip Architecture for Optimal Multi-Processor SoC Communication
    Rana, Vincenzo
    Atienza, David
    Santambrogio, Marco Domenico
    Sciuto, Donatella
    De Micheli, Giovanni
    VLSI-SOC: DESIGN METHODOLOGIES FOR SOC AND SIP, 2010, 313 : 232 - +
  • [39] Efficient Synchronization Methods for LET-based Applications on a Multi-Processor System on Chip
    Breaban, Gabriela
    Stuijk, Sander
    Goossens, Kees
    PROCEEDINGS OF THE 2017 DESIGN, AUTOMATION & TEST IN EUROPE CONFERENCE & EXHIBITION (DATE), 2017, : 1721 - 1726
  • [40] Eliminating migration in multi-processor scheduling
    Kalyanasundaram, B
    Pruhs, KR
    PROCEEDINGS OF THE TENTH ANNUAL ACM-SIAM SYMPOSIUM ON DISCRETE ALGORITHMS, 1999, : 499 - 506
12345