• DocumentCode
    2523765
  • Title

    Channel Interface: A Primitive Model for Memory Efficient Communication

  • Author

    Nanri, Takeshi ; Soga, Takeshi ; Ajima, Yuichiro ; Morie, Yoshiyuki ; Honda, Hiroaki ; Kobayashi, Taizo ; Takami, Toshiya ; Sumimoto, Shinji

  • Author_Institution
    Res. Inst. for Inf. Technol., Kyushu Univ., Fukuoka, Japan
  • fYear
    2015
  • fDate
    4-6 March 2015
  • Firstpage
    177
  • Lastpage
    181
  • Abstract
    Though the size of the system is getting larger towards exa-scale computation, the amount of available memory on computing nodes is expected to remain the same or to decrease. Therefore, memory efficiency is becoming an important issue for achieving scalability. This paper pointed out the problem of memory-inefficiency in the de-facto standard parallel programming model, Message Passing Interface (MPI). To solve this problem, the channel interface was introduced in the paper. This interface enables the programmers to appropriately allocate and de-allocate channels so that the program consumes just-enough amount of memory for communication. In addition to that, by limiting the message transfer supported by a channel as simple as possible, the memory consumption and the overhead for handling messages with this interface can be minimal. This paper showed a sample implementation of this interface. Then, the memory efficiency of the implementation is examined by the models of the memory consumption and the performance.
  • Keywords
    application program interfaces; message passing; parallel programming; MPI; channel allocation; channel deallocation; channel interface; computing nodes; de-facto standard parallel programming model; exa-scale computation; memory consumption; memory efficient communication; memory-inefficiency problem; message handling; message passing interface; message transfer; primitive model; Indexes; Libraries; Memory management; Message passing; Peer-to-peer computing; Protocols; Receivers;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Parallel, Distributed and Network-Based Processing (PDP), 2015 23rd Euromicro International Conference on
  • Conference_Location
    Turku
  • ISSN
    1066-6192
  • Type

    conf

  • DOI
    10.1109/PDP.2015.83
  • Filename
    7092717