Remote execution of data-parallel programs

被引:0
|
作者
Borowiec, J [1 ]
机构
[1] GMD FIRST, Res Inst Comp Architecture & Software Technol, Berlin, Germany
关键词
data mapping; data parallelism; server; client; stationary agent; HPF; CORBA; !text type='JAVA']JAVA[!/text; IDL; remote method call; adapter;
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
A framework is described that allows numerical computations to be performed in parallel on a collection of processors located anywhere on, a global network. The program to be executed is parallelized using data mapping mechanisms as defined e.g. in High Performance Fortran. Network: communication relies on the infrastructure provided by CORBA. The whole framework is implemented in Java. The implemented parallel execution scheme is as follows: Each of the cooperating processors hosts a server object that creates one or more stationary agents. Computation progress is supervised by a chosen server, the conductor, which is also a client. The conductor prompts the stationary agents to communicate and perform parallel computations in accordance with the program's control flow and data mapping.
引用
收藏
页码:1272 / 1279
页数:8
相关论文
共 50 条
  • [1] Combining fusion optimizations and piecewise execution of nested data-parallel programs
    Pfannenstiel, W
    [J]. PARALLEL AND DISTRIBUTED PROCESSING, PROCEEDINGS, 2000, 1800 : 324 - 331
  • [2] On privatization of variables for data-parallel execution
    Gupta, M
    [J]. 11TH INTERNATIONAL PARALLEL PROCESSING SYMPOSIUM, PROCEEDINGS, 1997, : 533 - 541
  • [3] Pipelined execution of data-parallel algorithms
    Gorev, Maksim
    Ubar, Raimund
    [J]. 2014 PROCEEDINGS OF THE 14TH BIENNIAL BALTIC ELECTRONICS CONFERENCE (BEC 2014), 2014, : 109 - 112
  • [4] SIZE AND ACCESS INFERENCE FOR DATA-PARALLEL PROGRAMS
    CHATTERJEE, S
    BLELLOCH, GE
    FISHER, AL
    [J]. SIGPLAN NOTICES, 1991, 26 (06): : 130 - 144
  • [5] Streaming networks for coordinating data-parallel programs
    Grelck, Clemens
    Scholz, Sven-Bodo
    Shafarenko, Alex
    [J]. PERSPECTIVES OF SYSTEMS INFORMATICS, 2007, 4378 : 451 - +
  • [6] Compiling data-parallel programs for clusters of SMPs
    Benkner, S
    Brandes, T
    [J]. CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE, 2004, 16 (2-3): : 111 - 132
  • [7] VISUALIZING THE PERFORMANCE OF SPMD AND DATA-PARALLEL PROGRAMS
    ROVER, DT
    WRIGHT, CT
    [J]. JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 1993, 18 (02) : 129 - 146
  • [8] Data-Parallel String-Manipulating Programs
    Veanes, Margus
    Mytkowicz, Todd
    Molnar, David
    Livshits, Benjamin
    [J]. ACM SIGPLAN NOTICES, 2015, 50 (01) : 139 - 152
  • [9] FPGA Circuit Synthesis of Accelerator Data-Parallel Programs
    Bond, Barry
    Hammil, Kerry
    Litchev, Lubomir
    Singh, Satnam
    [J]. 2010 18TH IEEE ANNUAL INTERNATIONAL SYMPOSIUM ON FIELD-PROGRAMMABLE CUSTOM COMPUTING MACHINES (FCCM 2010), 2010, : 167 - 170
  • [10] Automatic Insertion of Copy Annotation in Data-Parallel Programs
    Diniz Mendonca, Gleison Souza
    Ferreira Guimaraes, Breno Campos
    Oliveira Alves, Pericles Rafael
    Quintao Pereira, Fernando Magno
    Pereira, Marcio Machado
    Araujo, Guido
    [J]. PROCEEDINGS OF 28TH IEEE INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE AND HIGH PERFORMANCE COMPUTING, (SBAC-PAD 2016), 2016, : 34 - 41