Prospects of distributed shared memory for reducing global traffic in shared-bus multiprocessors

被引:0
|
作者
Lee, G
Kong, JS [1 ]
机构
[1] Univ Minnesota, Dept Comp Sci, Minneapolis, MN 55455 USA
[2] Univ Texas, Div Engn, San Antonio, TX 78249 USA
关键词
cache-only memory architecture; distributed shared-memory; non-uniform memory architecture; shared-bus multiprocessor;
D O I
暂无
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
As an effort, not a mutually exclusive but rather complementary to developing better backplane bus, this paper considers adapting distributed shared-memory (DSM) architectures to improve traditional shared-bus designs. We consider two well-known DSM architectures, namely Cache-coherent Non-Uniform Memory Architecture (NUMA) and Cache-Only Memory Architecture (COMA), in reducing bus traffic. Our study shows that COMA provides excellent opportunity of significantly reducing bandwidth requirement for the bus while cache-coherent NUMA provides a marginal improvement.
引用
收藏
页码:867 / 872
页数:6
相关论文
共 50 条
  • [41] An effective shared memory allocator for reducing false sharing in NUMA multiprocessors
    Lee, JW
    Cho, YK
    1996 IEEE SECOND INTERNATIONAL CONFERENCE ON ALGORITHMS & ARCHITECTURES FOR PARALLEL PROCESSING, ICA3PP'96, PROCEEDINGS OF, 1996, : 373 - 382
  • [42] A performance evaluation of cache injection in bus-based shared memory multiprocessors
    Milenkovic, A
    Milutinovic, V
    MICROPROCESSORS AND MICROSYSTEMS, 2002, 26 (02) : 51 - 61
  • [43] Distributed shared abstractions (DSA) on multiprocessors
    Clemencon, C
    Mukherjee, B
    Schwan, K
    IEEE TRANSACTIONS ON SOFTWARE ENGINEERING, 1996, 22 (02) : 132 - 152
  • [44] Design and implementation of an ABR server in a shared-bus ATM switch
    Zervanos, E
    Stassinopoulos, G
    ISCC 2000: FIFTH IEEE SYMPOSIUM ON COMPUTERS AND COMMUNICATIONS, PROCEEDINGS, 2000, : 366 - 371
  • [45] Exploration of distributed shared memory architectures for NoC-based multiprocessors
    Monchiero, Matteo
    Palermo, Gianluca
    Silvano, Cristina
    Villa, Oreste
    JOURNAL OF SYSTEMS ARCHITECTURE, 2007, 53 (10) : 719 - 732
  • [46] Exploration of distributed shared memory architectures for NoC-based multiprocessors
    Monchiero, Matteo
    Palermo, Gianluca
    Silvano, Cristina
    Villa, Oreste
    2006 INTERNATIONAL CONFERENCE ON EMBEDDED COMPUTER SYSTEMS: ARCHITECTURES, MODELING AND SIMULATION, PROCEEDINGS, 2006, : 144 - +
  • [47] Hybrid full map directory scheme for distributed shared memory multiprocessors
    Choi, JH
    Park, KH
    HIGH PERFORMANCE COMPUTING ON THE INFORMATION SUPERHIGHWAY - HPC ASIA '97, PROCEEDINGS, 1997, : 30 - 34
  • [48] Comparison of algorithms for transient stability simulations on shared and distributed memory multiprocessors
    LaScala, M
    Sblendorio, G
    Bose, A
    Wu, JQ
    IEEE TRANSACTIONS ON POWER SYSTEMS, 1996, 11 (04) : 2045 - 2050
  • [49] Chip-Level Redundancy in Distributed Shared-Memory Multiprocessors
    Gold, Brian T.
    Falsafi, Babak
    Hoe, Jarnes C.
    IEEE 15TH PACIFIC RIM INTERNATIONAL SYMPOSIUM ON DEPENDABLE COMPUTING, PROCEEDINGS, 2009, : 195 - +
  • [50] Hybrid full map directory scheme for distributed shared memory multiprocessors
    Korea Advanced Inst of Science and, Technology, Taejon, Korea, Republic of
    Proceedings of the Conference on High Performance Computing on the Information Superhighway, HPC Asia'97, 1997, : 30 - 34