On the Reproducibility of MPI Reduction Operations

被引:7
|
作者
Balaji, Pavan [1 ]
Kimpe, Dries [1 ]
机构
[1] Argonne Natl Lab, Math & Comp Sci Div, Argonne, IL 60439 USA
关键词
MPI; Reduction Operations; Reproducibility; Performance; Multicore; Topology;
D O I
10.1109/HPCC.and.EUC.2013.65
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Many scientific applications go through a thorough validation and verification ("V&V") process to demonstrate that the computer simulation does, in fact, mirror what can be analyzed through physical experimentation. Given the complexity of and the time required for the V&V process, applications that have been validated and verified are typically conservative with respect to changes that might impact the reproducibility of their results. In the extreme case, some applications require bitwise reproducibility for their simulations. Thus, any change made to the application, the hardware, or the software on the system needs to preserve the bitwise reproducibility of the application. Such a constraint, however, can drastically affect the performance efficiency of the system in many ways. In this paper, we analyze the impact of such bitwise reproducibility on the performance efficiency of MPI reduction operations.
引用
收藏
页码:407 / 414
页数:8
相关论文
共 50 条
  • [1] A Statistical Analysis of Error in MPI Reduction Operations
    Pollard, Samuel D.
    Norris, Boyana
    PROCEEDINGS OF CORRECTNESS 2020: FOUTH IEEE/ACM INTERNATIONAL WORKSHOP ON SOFTWARE CORRECTNESS FOR HPC APPLICATIONS, 2020, : 49 - 57
  • [2] Transparent Neutral Element Elimination in MPI Reduction Operations
    Traeff, Jesper Larsson
    RECENT ADVANCES IN THE MESSAGE PASSING INTERFACE, 2010, 6305 : 275 - 284
  • [3] MPI Reduction Operations for Sparse Floating-point Data
    Hofmann, Michael
    Ruenger, Gudula
    RECENT ADVANCES IN PARALLEL VIRTUAL MACHINE AND MESSAGE PASSING INTERFACE, 2008, 5205 : 94 - 101
  • [4] Sparse Collective Operations for MPI
    Hoefler, Torsten
    Traeff, Jesper Larsson
    2009 IEEE INTERNATIONAL SYMPOSIUM ON PARALLEL & DISTRIBUTED PROCESSING, VOLS 1-5, 2009, : 1659 - +
  • [5] On benchmarking collective MPI operations
    Worsch, T
    Reussner, R
    Augustin, W
    RECENT ADVANCES IN PARALLEL VIRTUAL MACHINE AND MESSAGE PASSING INTERFACE, PROCEEDINGS, 2002, 2474 : 271 - 279
  • [6] Automatic Partitioning of MPI Operations in MPI plus OpenMP Applications
    Jammer, Tim
    Bischof, Christian
    HIGH PERFORMANCE COMPUTING - ISC HIGH PERFORMANCE DIGITAL 2021 INTERNATIONAL WORKSHOPS, 2021, 12761 : 191 - 198
  • [7] Pattern matching of collective MPI operations
    Kranzlmüller, D
    Knüpfer, A
    Nagel, WE
    PDPTA '04: PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED PROCESSING TECHNIQUES AND APPLICATIONS, VOLS 1-3, 2004, : 1243 - 1249
  • [8] Pipelining and overlapping for MPI collective operations
    Worringen, J
    LCN 2003: 28TH CONFERENCE ON LOCAL COMPUTER NETWORKS, PROCEEDINGS, 2003, : 548 - 557
  • [9] Performance analysis of MPI collective operations
    Pjesivac-Grbovic, Jelena
    Angskun, Thara
    Bosilca, George
    Fagg, Graham E.
    Gabriel, Edgar
    Dongarra, Jack J.
    CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS, 2007, 10 (02): : 127 - 143
  • [10] Performance analysis of MPI collective operations
    Pješivac-Grbović, Jelena
    Angskun, Thara
    Bosilca, George
    Fagg, Graham E.
    Gabriel, Edgar
    Dongarra, Jack J.
    Cluster Computing, 2007, 10 (02) : 127 - 143