Using Triggered Operations to Offload Collective Communication Operations

被引:0
|
作者
Hemmert, K. Scott [1 ]
Barrett, Brian [1 ]
Underwood, Keith D. [2 ]
机构
[1] Sandia Natl Labs, POB 5800,MS-1110, Albuquerque, NM 87185 USA
[2] INTEL Corp, Hillsboro, OR USA
关键词
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
Efficient collective operations are a major component of application scalability. Offload of collective operations onto the network interface reduces many of the latencies that are inherent in network communications and, consequently, reduces the time to perform the collective operation. To support offload, it is desirable to expose semantic building blocks that are simple to offload and yet powerful enough to implement a variety of collective algorithms. This paper presents the implementation of barrier and broadcast leveraging triggered operations a semantic building block for collective offload. Triggered operations are shown to be both semantically powerful and capable of improving performance.
引用
收藏
页码:249 / +
页数:2
相关论文
共 50 条
  • [1] Using Triggered Operations to Offload Rendezvous Messages
    Barrett, Brian W.
    Brightwell, Ron
    Hemmert, K. Scott
    Wheeler, Kyle B.
    Underwood, Keith D.
    RECENT ADVANCES IN THE MESSAGE PASSING INTERFACE, 2011, 6960 : 120 - +
  • [2] Minimizing the usage of hardware counters for collective communication using triggered operations
    Islam, Nusrat Sharmin
    Zheng, Gengbin
    Sur, Sayantan
    Langer, Akhil
    Garzaran, Maria
    PARALLEL COMPUTING, 2020, 98
  • [3] Minimizing the Usage of Hardware Counters for Collective Communication using Triggered Operations
    Islam, Nusrat Sharmin
    Zheng, Gengbin
    Sur, Sayantan
    Langer, Akhil
    Garzaran, Maria
    EUROMPI'19: PROCEEDINGS OF THE 26TH EUROPEAN MPI USERS' GROUP MEETING, 2019,
  • [4] Optimization of collective communication operations in MPICH
    Thakur, R
    Rabenseifner, R
    Gropp, W
    INTERNATIONAL JOURNAL OF HIGH PERFORMANCE COMPUTING APPLICATIONS, 2005, 19 (01): : 49 - 66
  • [5] Kernel-Based Offload of Collective Operations - Implementation, Evaluation and Lessons Learned
    Schneider, Timo
    Eckelmann, Sven
    Hoefler, Torsten
    Rehm, Wolfgang
    EURO-PAR 2011 PARALLEL PROCESSING, PT 2, 2011, 6853 : 264 - 275
  • [6] EXPERIMENTAL RESULTS ABOUT MPI COLLECTIVE COMMUNICATION OPERATIONS
    Bernaschi, Massimo
    Iannello, Giulio
    Crea, Saverio
    PARALLEL PROCESSING LETTERS, 2005, 15 (1-2)
  • [7] Validation of dimemas communication model for MPI collective operations
    Girona, S
    Labarta, J
    Badia, RM
    RECENT ADVANCES IN PARALLEL VIRTUAL MACHINE AND MESSAGE PASSING INTERFACE, PROCEEDINGS, 2000, 1908 : 39 - 46
  • [8] ECO: Efficient Collective Operations for communication on heterogeneous networks
    Lowekamp, BB
    Beguelin, A
    10TH INTERNATIONAL PARALLEL PROCESSING SYMPOSIUM - PROCEEDINGS OF IPPS '96, 1996, : 399 - 405
  • [9] Experimental results about MPI collective communication operations
    Bernaschi, M
    Iannello, G
    Lauria, M
    HIGH-PERFORMANCE COMPUTING AND NETWORKING, PROCEEDINGS, 1999, 1593 : 774 - 783
  • [10] An adaptive extension library for improving collective communication operations
    Hartmann, O.
    Kuehnemann, M.
    Rauber, T.
    Ruenger, G.
    CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE, 2008, 20 (10): : 1173 - 1194