PseudoShare: Bring Shared to Exclusive for Main Memory in Multi-core Systems

被引:0
|
作者
Meng, Xiaolin [1 ,2 ]
Jia, Gangyong [1 ,2 ]
Wan, Jian [1 ,2 ]
Zhang, Jilin [1 ,2 ]
机构
[1] Hangzhou Dianzi Univ, Dept Comp Sci & Technol, Hangzhou 310018, Zhejiang, Peoples R China
[2] Hangzhou Dianzi Univ, Key Lab Complex Syst Modeling & Simulat, Minist Educ, Hangzhou 310018, Zhejiang, Peoples R China
关键词
Memory contention; memory interference; performance degradation differently; unfairness resouce sharing; power;
D O I
10.1109/CSE.2014.347
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
In modern multi-core system, memory is shared among more and more concurrently running threads. Therefore, memory contention and interference is more and more seriously which induces performance degradation unevenly, unfairness resource sharing and priority inversion even starvation. In this paper, we first analyze the problems induced by contention and interference in detail, and then, propose pseudoshare framework which brings shared memory to exclusive in multi-core system. The pseudoshare framework contains three steps: 1) Partition threads and memory into thread and memory groups respectively, each thread group runs on one core occupying one memory group; 2) Analyze the performance influence after memory shared for each thread; 3) According to performance influence of parallel running threads, allocate memory bandwidth for each thread. We implement pseudoshare in 4-core and 8-core platforms. Experimental results show pseudoshare optimizes both problems of performance degradation differently and unfairness resource sharing in both 4-core and 8-core. Moreover, pseudoshare solves starvation and reduces 5.3% power on average.
引用
收藏
页码:1894 / 1899
页数:6
相关论文
共 50 条
  • [1] Novel parallel method for association rule mining on multi-core shared memory systems
    Vu, Lan
    Alaghband, Gita
    [J]. PARALLEL COMPUTING, 2014, 40 (10) : 768 - 785
  • [2] Improving the scalability of shared cache multi-core systems
    Prabhu, Sapna
    Daruwala, R. D.
    [J]. 2014 ANNUAL IEEE INDIA CONFERENCE (INDICON), 2014,
  • [3] Task Assignments based on Shared Memory Multi-core Communication
    Xu, Xiaojie
    Wang, Lisheng
    [J]. 2014 2ND INTERNATIONAL CONFERENCE ON SYSTEMS AND INFORMATICS (ICSAI), 2014, : 324 - 328
  • [4] Study on multi-core data structure in shared-memory
    Zhou W.
    Zhou K.-R.
    Luan Z.-Z.
    Yao S.-W.
    Qian D.-P.
    [J]. Zhou, Wei (zwei@ynu.edu.cn), 2016, Chinese Academy of Sciences (27): : 1009 - 1025
  • [5] Decoupled MapReduce for Shared-Memory Multi-Core Architectures
    Iliakis, Konstantinos
    Xydis, Sotirios
    Soudris, Dimitrios
    [J]. IEEE COMPUTER ARCHITECTURE LETTERS, 2018, 17 (02) : 143 - 146
  • [6] Hybrid Main Memory for High Bandwidth Multi-Core System
    Kim, Dongki
    Yoo, Sungjoo
    Lee, Sunggu
    [J]. IEEE TRANSACTIONS ON MULTI-SCALE COMPUTING SYSTEMS, 2015, 1 (03): : 138 - 149
  • [7] Main-Memory Scan Sharing For Multi-Core CPUs
    Qiao, Lin
    Raman, Vijayshankar
    Reiss, Frederick
    Haas, Peter J.
    Lohman, Guy M.
    [J]. PROCEEDINGS OF THE VLDB ENDOWMENT, 2008, 1 (01): : 610 - 621
  • [8] ANALYSIS OF THE EFFICIENCY OF ATOMIC OPERATIONS IN MULTI-CORE SHARED-MEMORY COMPUTER SYSTEMS.
    Goncharenko, E. A.
    Paznikov, A. A.
    [J]. VESTNIK TOMSKOGO GOSUDARSTVENNOGO UNIVERSITETA-UPRAVLENIE VYCHISLITELNAJA TEHNIKA I INFORMATIKA-TOMSK STATE UNIVERSITY JOURNAL OF CONTROL AND COMPUTER SCIENCE, 2020, (51): : 102 - 110
  • [9] A Multi-core Memory Organization for 3-D DRAM as Main Memory
    Sherman, Jared
    Kavi, Krishna
    Potter, Brandon
    Ignatowski, Mike
    [J]. ARCHITECTURE OF COMPUTING SYSTEMS - ARCS 2013, 2013, 7767 : 62 - 73
  • [10] Parallel finite element solver for multi-core computers with shared memory
    Fialko, Sergiy
    [J]. COMPUTERS & MATHEMATICS WITH APPLICATIONS, 2021, 94 : 1 - 14