ParaStation: Efficient parallel computing by clustering workstations: Design and evaluation

被引:0
|
作者
Warschko, TM [1 ]
Blum, JM [1 ]
Tichy, WF [1 ]
机构
[1] Univ Karlsruhe, Dept Informat, D-76128 Karlsruhe, Germany
关键词
workstation cluster; parallel and distributed computing; user-level communication; high-speed interconnects;
D O I
暂无
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
ParaStation is a communications fabric for connecting off-the-shelf workstations into a supercomputer. The fabric employs technology used in massively parallel machines and scales up to 4096 nodes, ParaStation's user-level message passing software preserves the low latency of the fabric by taking the operating system out of the communication path, while still providing full protection in a multiprogramming environment. The programming interface presented by ParaStation consists of a UNIX socket emulation and widely used parallel programming environments such as PVM, P4, and MPI. Implementations of ParaStation using various platforms, such as Digitals AlphaGeneration workstations and Linux PCs, achieve end-to-end (process-to-process) latencies as low as 2 mu s and a sustained bandwidth of up to 15 Mbyte/s per channel, even with small packets. Benchmarks using PVM on ParaStation demonstrate real application performance of 1 GFLOP on an 8-node cluster.
引用
收藏
页码:241 / 260
页数:20
相关论文
共 50 条
  • [41] Clustering Weighted Slope One for Distributed Parallel Computing
    Xing Chen
    Wang Hongfa
    2011 INTERNATIONAL CONFERENCE ON COMPUTER SCIENCE AND NETWORK TECHNOLOGY (ICCSNT), VOLS 1-4, 2012, : 1595 - 1598
  • [42] Heterogeneous Clustering Computing Based on Parallel Task Distribution
    Lv, Yi
    Chen, Wenshi
    NETWORK COMPUTING AND INFORMATION SECURITY, 2012, 345 : 485 - +
  • [43] Design model for assembly lines including fractional tasks and parallel workstations
    Renna, Paolo
    INTERNATIONAL JOURNAL OF INTERACTIVE DESIGN AND MANUFACTURING - IJIDEM, 2024, 18 (01): : 177 - 190
  • [44] Design model for assembly lines including fractional tasks and parallel workstations
    Paolo Renna
    International Journal on Interactive Design and Manufacturing (IJIDeM), 2024, 18 : 177 - 190
  • [45] Implementation and performance evaluation of parallel computations by using cluster of networked workstations
    Wang, H
    Prabhu, GM
    Takle, ES
    Shen, J
    PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED PROCESSING TECHNIQUES AND APPLICATIONS, VOLS I-V, 2000, : 2007 - 2013
  • [46] Efficient sorting design on a novel embedded parallel computing architecture with unique memory access
    Zhou, Wenbiao
    Cai, Zhaoyun
    Ding, Ruiqiang
    Gong, Chen
    Liu, Dake
    COMPUTERS & ELECTRICAL ENGINEERING, 2013, 39 (07) : 2100 - 2111
  • [47] Fast and Efficient Parallel Computations Using a Cluster of Workstations to Simulate Flood Flows
    Mungkasi, Sudi
    Darmawan, J. B. Budi
    INTELLIGENCE IN THE ERA OF BIG DATA, ICSIIT 2015, 2015, 516 : 469 - 477
  • [48] Implementation and evaluation of parallel edge detector on cluster of workstations using MPI
    Patel, N
    Mahgoub, IO
    10TH INTERNATIONAL CONFERENCE ON COMPUTER APPLICATIONS IN INDUSTRY AND ENGINEERING, 1997, : 102 - 105
  • [49] Efficient Quality Threshold Clustering for Parallel Architectures
    Danalis, Anthony
    McCurdy, Collin
    Vetter, Jeffrey S.
    2012 IEEE 26TH INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM (IPDPS), 2012, : 1068 - 1079
  • [50] An efficient clustering algorithm for partitioning parallel programs
    Maheshwari, P
    Shen, H
    PARALLEL COMPUTING, 1998, 24 (5-6) : 893 - 909