EXPLOITING OPERATING SYSTEM SUPPORT FOR DYNAMIC PAGE PLACEMENT ON A NUMA SHARED MEMORY MULTIPROCESSOR

被引:0
|
作者
LAROWE, RP [1 ]
WILKES, JT [1 ]
ELLIS, CS [1 ]
机构
[1] DUKE UNIV,DEPT COMP SCI,DURHAM,NC 27706
来源
SIGPLAN NOTICES | 1991年 / 26卷 / 07期
关键词
D O I
暂无
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
Shared memory multiporcessors are attractive because they are programmed in a manner similar to uniprocessors. The UMA class of shared memory multiprocessors is the most attractive, from the programmer's point of view, since the programmer need not be concerned with the placement of code and data in the physical memory hierarchy. Scalable shared memory multiprocessors, on the other hand, tend to present at least some degree of non-uniformity of memory access to the programmer, making the NUMA class an important one to consider. In this paper, we investigate the role that DUnX, an operating system supporting dynamic page placement on a BBN GP1000, might play in simplifying the memory model presented to the applications programmer. We consider a case study of psolu, a real scientific application originally targeted for a NUMA architecture. We find that dynamic page placement can dramatically improve the performance of a simpler implementation of psolu targeted for an UMA memory architecture. The addition of a phase change hint to the UMA version of psolu enables the operating system to further improve performance, approaching that of the hand-tuned NUMA implementation.
引用
收藏
页码:122 / 132
页数:11
相关论文
共 50 条
  • [31] Experiences from implementing multiprocessor support for an industrial operating system kernel
    Kågström, S
    Grahn, H
    Lundberg, L
    11TH IEEE INTERNATIONAL CONFERENCE ON EMBEDDED AND REAL-TIME COMPUTING SYSTEMS AND APPLICATIONS, PROCEEDINGS, 2005, : 365 - 368
  • [32] Integrating dynamic memory placement with adaptive load-balancing for parallel codes on NUMA multiprocessors
    Slavin, Paul
    Freeman, Len
    EURO-PAR 2008 PARALLEL PROCESSING, PROCEEDINGS, 2008, 5168 : 274 - 284
  • [33] Reduction of false sharing by using process affinity in page-based distributed shared memory multiprocessor systems
    Hung, KP
    Yung, NHC
    Cheung, YS
    1996 IEEE SECOND INTERNATIONAL CONFERENCE ON ALGORITHMS & ARCHITECTURES FOR PARALLEL PROCESSING, ICA3PP'96, PROCEEDINGS OF, 1996, : 383 - 390
  • [34] Architectural and Operating System Support for Virtual Memory
    Bhattacharjee A.
    Lustig D.
    1600, Morgan and Claypool Publishers (12): : 1 - 177
  • [35] A survey of operating system support for persistent memory
    Miao CAI
    Hao HUANG
    Frontiers of Computer Science, 2021, (04) : 13 - 32
  • [36] A survey of operating system support for persistent memory
    Miao Cai
    Hao Huang
    Frontiers of Computer Science, 2021, 15
  • [37] A survey of operating system support for persistent memory
    Cai, Miao
    Huang, Hao
    FRONTIERS OF COMPUTER SCIENCE, 2021, 15 (04)
  • [38] Low power tasks mapping for DVS capable multiprocessor system with shared memory
    Li, M
    Wu, XB
    Zhu, XH
    Wang, H
    PROCEEDINGS OF THE IASTED INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED COMPUTING AND NETWORKS, 2004, : 126 - 133
  • [39] Predicting the performance measures of an optical distributed shared memory multiprocessor by using support vector regression
    Akay, M. Fatih
    Abasikeles, Ipek
    EXPERT SYSTEMS WITH APPLICATIONS, 2010, 37 (09) : 6293 - 6301
  • [40] Dynamic thread mapping of shared memory applications by exploiting cache coherence protocols
    Cruz, Eduardo H. M.
    Diener, Matthias
    Alves, Marco A. Z.
    Navaux, Philippe O. A.
    JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 2014, 74 (03) : 2215 - 2228