Architecture and the software environment of parallel computer Cenju-4

被引:0
|
作者
Nakata, T [1 ]
Kanoh, Y
Tatsukawa, K
Yanagida, S
Nishi, N
Takayama, H
机构
[1] NEC Corp Ltd, C&C Media Res Labs, Tokyo, Japan
[2] NEC Corp Ltd, Informatec Syst, Tokyo, Japan
[3] NEC Corp Ltd, Comp Div, Tokyo, Japan
来源
NEC RESEARCH & DEVELOPMENT | 1998年 / 39卷 / 04期
关键词
high performance computing (HPC); parallel processing; distributed memory architecture; distributed shared memory (DSM);
D O I
暂无
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
This paper describes the architecture and operating system, and gives an evaluation of NEC's new parallel computer Cenju-4 Major features of Cenju-4 are: a) parallel memory architecture which encompasses distributed shared memory and user-level inter-processor communication. b) Scalable system from 8 nodes to 1,024 nodes. Using the powerful RISC processor VR10000 (200 MHz) from MIPS II Technologies, Inc., Cenju-4 system can be configured from 8 nodes to 1,024 nodes, flexibly extending the system as the demand arises. c) Utilization of a flexible micro-kernel operating system. Since the system adopts a micro-kernel based operating system (MACH), it can be configured into several software environments such as UNIX double dagger server systems and, single system image systems. The key components of the system are two 1 M gate arrays which implement memory control, inter processor communication control and network communication controls. The programming environment provided are de-facto standard libraries, high-level programming languages such as MPI (Message Passing Interface), PVM (Parallel Virtual Machine) and HPF (High Performance Fortran). The operating system and the inter-processor communication libraries fully exploit the functionality of the hardware to realize an inter-processor communication latency of 4.5 mu s and the throughput of 169 MB/s at user program level.
引用
收藏
页码:385 / 390
页数:6
相关论文
共 50 条
  • [1] Architecture of a parallel computer Cenju-4
    Kanoh, Y
    Hirose, T
    Nakamura, M
    Hosomi, T
    Tatsukawa, K
    Araki, H
    Sugawara, T
    Nakata, T
    [J]. INNOVATIVE ARCHITECTURE FOR FUTURE GENERATION HIGH-PERFORMANCE PROCESSORS AND SYSTEMS, 1999, : 105 - 113
  • [2] The Omni OpenMP compiler on the distributed shared memory of Cenju-4
    Kusano, K
    Sato, M
    Hosomi, T
    Seo, Y
    [J]. OPENMP SHARED MEMORY PARALLEL PROGRAMMING, PROCEEDINGS, 2001, 2104 : 20 - 30
  • [3] Architecture of a parallel machine: Cenju3
    Maruyama, T
    Kanoh, Y
    Hirose, T
    Muramatsu, K
    Nakata, T
    Asano, Y
    Inamura, Y
    [J]. SYSTEMS AND COMPUTERS IN JAPAN, 1995, 26 (14) : 26 - 36
  • [4] A SOFTWARE ENVIRONMENT FOR PARALLEL COMPUTER VISION
    JAMIESON, LH
    DELP, EJ
    WANG, CC
    LI, J
    WEIL, FJ
    [J]. COMPUTER, 1992, 25 (02) : 73 - 77
  • [5] FreeTIV parallel computer: Architecture and environment
    Amiot, F
    Pissaloux, EE
    [J]. 5TH INTERNATIONAL WORKSHOP ON COMPUTER ARCHITECTURES FOR MACHINE PERCEPTION, PROCEEDINGS, 2000, : 12 - 17
  • [6] FreeTIV parallel computer: Architecture and environment
    [J]. Amiot, Franck, 2000, IEEE, Piscataway, NJ, United States
  • [7] Parallel computer architecture
    Müller, S
    Stenström, P
    Valero, M
    Vassiliadis, S
    [J]. EURO-PAR 2000 PARALLEL PROCESSING, PROCEEDINGS, 2000, 1900 : 537 - 538
  • [8] A UNIVERSAL PARALLEL COMPUTER ARCHITECTURE
    DALLY, WJ
    [J]. NEW GENERATION COMPUTING, 1993, 11 (3-4) : 227 - 249
  • [9] Software architecture for evolving environment
    Kral, Jaroslav
    Zemlicka, Michal
    [J]. 13TH IEEE INTERNATIONAL WORKSHOP ON SOFTWARE TECHNOLOGY AND ENGINEERING PRACTICE, PROCEEDINGS, 2006, : 49 - +
  • [10] Parallel evaluation of software architecture specifications
    Tsai, JJP
    Li, B
    Juan, EYT
    [J]. COMMUNICATIONS OF THE ACM, 1997, 40 (01) : 83 - 86