COMPUTATION ON PARALLEL MESSAGE-PASSING COMPUTERS

被引:0
|
作者
PETERSEN, J
机构
[1] Bergen Scientific Centre, IBM, Thormøhlens gt, Bergen
来源
PHYSICA SCRIPTA | 1991年 / T38卷
关键词
D O I
10.1088/0031-8949/1991/T38/006
中图分类号
O4 [物理学];
学科分类号
0702 ;
摘要
Within parallel computing, one may divide the present architectures into 3 categories, (a) massively parallel SIMD, (b) Shared memory MIMD and (c) Distributed memory MIMD (or Message Passing). (SIMD = Single Instruction Multiple Data; MIMD = Multiple Instruction Multiple Data). The “simplest” of these is the message-passing architecture. A Message-Passing system consists of several processors, where each processor has a “fair” amount of local memory and is connected to some kind of communication network. This talk will cover some of the most common tools of an established parallel message passing operating system, Trollius [1]. One tool, TGraph [2], developed at Bergen Scientific Centre, is a tool for graphics output under X-windows from the processors. A programming philosophy [3] for message-passing computers which is based on ‘domain decomposition’, and some specific applications, such as Wave Equation [4], Gauss Elimination [5] and a large Climate Model project will be discussed. © 1991 IOP Publishing Ltd.
引用
收藏
页码:33 / 33
页数:1
相关论文
共 50 条
  • [1] COMPETITIVE NEURAL NETWORKS ON MESSAGE-PASSING PARALLEL COMPUTERS
    CECCARELLI, M
    PETROSINO, A
    VACCARO, R
    CONCURRENCY-PRACTICE AND EXPERIENCE, 1993, 5 (06): : 449 - 470
  • [2] Processor allocation policies for message-passing parallel computers
    McCann, Cathy
    Zahorjan, John
    Performance Evaluation Review, 1994, 22 (01): : 19 - 32
  • [3] Bounds on the efficiency of message-passing protocols for parallel computers
    Cypher, R
    Konstantinidou, S
    SIAM JOURNAL ON COMPUTING, 1996, 25 (05) : 1082 - 1104
  • [4] Multiprecision matrix arithmetic for parallel computers with message-passing interface
    Morozov, VA
    Vazhenin, AP
    PROGRAMMING AND COMPUTER SOFTWARE, 1999, 25 (01) : 56 - 64
  • [5] GLOBAL COMMUNICATIONS ON LOCALLY CONNECTED MESSAGE-PASSING PARALLEL COMPUTERS
    DEBBAGE, M
    HILL, MB
    NICOLE, DA
    CONCURRENCY-PRACTICE AND EXPERIENCE, 1993, 5 (06): : 491 - 509
  • [6] MULTICOMPUTERS - MESSAGE-PASSING CONCURRENT COMPUTERS
    ATHAS, WC
    SEITZ, CL
    COMPUTER, 1988, 21 (08) : 9 - 24
  • [7] Message-passing performance of various computers
    Dongarra, JJ
    Dunigan, T
    CONCURRENCY-PRACTICE AND EXPERIENCE, 1997, 9 (10): : 915 - 926
  • [8] Finite element message-passing/DSM simulation algorithm for parallel computers
    Plazek, J
    Banas, K
    Kitowski, J
    HIGH-PERFORMANCE COMPUTING AND NETWORKING, 1998, 1401 : 878 - 880
  • [9] Efficient parallel prefix algorithms on fully connected message-passing computers
    Lin, YC
    Lin, CM
    3RD INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE COMPUTING, PROCEEDINGS, 1996, : 316 - 321
  • [10] The parallel 'Deutschland-Modell ' - A message-passing version for distributed memory computers
    Schattler, U
    Krenzien, E
    PARALLEL COMPUTING, 1997, 23 (14) : 2215 - 2226