A new relaxed memory consistency model for shared-memory multiprocessors with parallel-multithreaded processing elements

被引:0
|
作者
Wu, CC [1 ]
Chen, C [1 ]
机构
[1] Natl Chiao Tung Univ, Dept Comp Sci & Informat Engn, Hsinchu 300, Taiwan
关键词
memory consistency model; multithread; multiprocessor; write cache; synchronization; PSC model; barrier; performance evaluation;
D O I
暂无
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The release consistency model is the generally accepted hardware-centric relaxed memory consistency model because of its performance and implementation complexity. By extending the release consistency model, in this paper. we propose a hardware-centric memory consistency model particularly for shared-memory multiprocessor systems with parallel-multithreaded processing elements. The new model uses a new categorization for memory references and utilizes the feature of parallel multithreaded processors (PMPs). We further partition acquire and release references into three sub-categories: one for lock-unlock pairs, one for barrier synchronization, and the last for others. According to the semantic of each synchronization primitive, each sub-category has its own relaxed restrictions. On the other hand, the feature of a PMP is that it is capable of executing more than one thread at the same time, where all parallel threads share only one cache hierarchy. Under the new model, we can use dual write-caches to reduce write traffic and synchronization time. We have used five benchmarks in the SPLASH suite to evaluate the performance gain for the new model. According to the simulation results, the new model is superior to the release consistency model at best by about 11%.
引用
收藏
页码:785 / 808
页数:24
相关论文
共 50 条
  • [41] A MODEL FOR ASYNCHRONOUS SHARED-MEMORY PARALLEL COMPUTATION
    NISHIMURA, N
    SIAM JOURNAL ON COMPUTING, 1994, 23 (06) : 1231 - 1252
  • [42] Shared-memory multiprocessors: SW or HW support?
    Scott, S
    THIRD INTERNATIONAL SYMPOSIUM ON HIGH-PERFORMANCE COMPUTER ARCHITECTURE - PROCEEDINGS, 1997, : 140 - 140
  • [43] ALGORITHMS FOR SCALABLE SYNCHRONIZATION ON SHARED-MEMORY MULTIPROCESSORS
    MELLORCRUMMEY, JM
    SCOTT, ML
    ACM TRANSACTIONS ON COMPUTER SYSTEMS, 1991, 9 (01): : 21 - 65
  • [44] Cache Coherence Protocols in Shared-Memory Multiprocessors
    Lian, Xiuzhen
    Ning, Xiaoxi
    Xie, Mingren
    Yu, Farong
    PROCEEDINGS OF THE 2015 INTERNATIONAL CONFERENCE ON COMPUTATIONAL SCIENCE AND ENGINEERING, 2015, 17 : 286 - 289
  • [45] Design and analysis of algorithms for shared-memory multiprocessors
    Leiserson, CE
    ALGORITHMS AND DATA STRUCTURES, 1999, 1663 : 55 - 55
  • [46] Data forwarding in scalable shared-memory multiprocessors
    Koufaty, DA
    Chen, XF
    Poulsen, DK
    Torrellas, J
    IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 1996, 7 (12) : 1250 - 1264
  • [47] LANGUAGE PORTABILITY ACROSS SHARED-MEMORY MULTIPROCESSORS
    ALAGHBAND, G
    BENTEN, MS
    JAKOB, R
    JORDAN, HF
    RAMANAN, AV
    IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 1993, 4 (09) : 1064 - 1072
  • [48] Dynamic scheduling strategies for shared-memory multiprocessors
    Hamidzadeh, B
    Lilja, DJ
    PROCEEDINGS OF THE 16TH INTERNATIONAL CONFERENCE ON DISTRIBUTED COMPUTING SYSTEMS, 1996, : 208 - 215
  • [49] Coarse-grained thread pipelining: A speculative parallel execution model for shared-memory multiprocessors
    Kazi, IH
    Lija, DJ
    IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2001, 12 (09) : 952 - 966
  • [50] MEMORY MANAGEMENT FOR PARALLEL TASKS IN SHARED-MEMORY
    LANGENDOEN, KG
    MULLER, HL
    VREE, WG
    LECTURE NOTES IN COMPUTER SCIENCE, 1992, 637 : 165 - 178