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
  • [21] EXPERIMENTAL RESULTS ABOUT MPI COLLECTIVE COMMUNICATION OPERATIONS
    Bernaschi, Massimo
    Iannello, Giulio
    Crea, Saverio
    PARALLEL PROCESSING LETTERS, 2005, 15 (1-2)
  • [22] A Study of Process Arrival Patterns for MPI Collective Operations
    Faraj, Ahmad
    Patarasuk, Pitch
    Yuan, Xin
    INTERNATIONAL JOURNAL OF PARALLEL PROGRAMMING, 2008, 36 (06) : 543 - 570
  • [23] A Study of Process Arrival Patterns for MPI Collective Operations
    Ahmad Faraj
    Pitch Patarasuk
    Xin Yuan
    International Journal of Parallel Programming, 2008, 36 : 543 - 570
  • [24] Compiler-enabled optimization of persistent MPI Operations
    Jammer, Tim
    Bischof, Christian
    2022 IEEE/ACM INTERNATIONAL WORKSHOP ON EXASCALE MPI (EXAMPI), 2022, : 1 - 10
  • [25] Experimental results about MPI collective communication operations
    Bernaschi, M
    Iannello, G
    Lauria, M
    HIGH-PERFORMANCE COMPUTING AND NETWORKING, PROCEEDINGS, 1999, 1593 : 774 - 783
  • [26] Reproducibility of projection sieving operations with different frequencies
    Meyer, HJ
    AMERICAN LABORATORY, 2001, 33 (23) : 28 - 29
  • [27] Verification of halting properties for MPI programs using nonblocking operations
    Siegel, Stephen F.
    Avrunin, George S.
    RECENT ADVANCES IN PARALLEL VIRTUAL MACHINE AND MESSAGE PASSING INTERFACE, 2007, 4757 : 326 - +
  • [28] Adaptive MPI collective operations based on evaluations in LogP model
    Paznikov, A. A.
    Kupriyanov, M. S.
    14TH INTERNATIONAL SYMPOSIUM INTELLIGENT SYSTEMS, 2021, 186 : 323 - 330
  • [29] Hybrid MPI-thread parallelization of adaptive mesh operations
    Ibanez, Dan
    Dunn, Ian
    Shephard, Mark S.
    PARALLEL COMPUTING, 2016, 52 : 133 - 143
  • [30] COOL: A Cloud-Optimized Structure for MPI Collective Operations
    Alfatafta, Mohammed
    AlSader, Zuhair
    Al-Kiswany, Samer
    PROCEEDINGS 2018 IEEE 11TH INTERNATIONAL CONFERENCE ON CLOUD COMPUTING (CLOUD), 2018, : 746 - 753