Reducing Manipulation Overhead of Remote Data-Structure by Controlling Remote Memory Access Order

  • Yuichiro AjimaEmail author
  • Takafumi Nose
  • Kazushige Saga
  • Naoyuki Shida
  • Shinji Sumimoto
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9945)


The Advanced Communication Primitives (ACP) is a communication library which provides the PGAS programming model to existing programming languages. The communication primitives of ACP include remote-to-remote data transfer and atomic operations. The reference implementation of communication primitives of ACP uses connectionless sockets over UDP and agent threads. The remote-to-remote data transfer is implemented as a protocol. The ACP data library (ACPdl) is a utility library using the communication primitives that include interfaces to create and manipulate several types of remote and distributed data structures. In the current implementation of ACP, there is a performance issue in the erase and insert functions of vector-type data structures due to the in-place data movement algorithm. This paper proposes a new technique called ‘remote ordering’ for the remote-to-remote data transfer protocol. The remote ordering technique overlaps the progresses of the protocol for the data movement simultaneously. The evaluation results show that the average execution times of the functions were reduced to about one seventh.


Communication library Partitioned global address space Data structure 



The development of the ACP library is a part of the Advanced Communication for Exa (ACE) project, [15] which is a research theme in the CREST research area ‘Development of System Software Technologies for post-Peta Scale High Performance Computing,’ sponsored by JST (Japan Science and Technology Agency).


  1. 1.
    Sumimoto, S., Ajima, Y., Saga, K., Nose, T., Shida, N., Nanri, T.: The design of advanced communication to reduce memory usage for exa-scale systems. In: 12th International Meeting on High Performance Computing for Computational Science (VECPAR) (2016, in press)Google Scholar
  2. 2.
    Ajima, Y., Nose, T., Saga, K., Shida, N., Sumimoto, S.: ACPdl: data-structure and global memory allocator library over a thin PGAS-layer. In: The First International Workshop on Extreme Scale Programming Models and Middleware (ESPM 2015), pp. 11–18 (2015)Google Scholar
  3. 3.
    Nanri, T., Soga, T., et al.: Channel interface: a primitive model for memory efficient communication. In: 23rd Euromicro PDP, pp. 177–181 (2015)Google Scholar
  4. 4.
    Postel, J.B. (ed.): User Datagram Protocol, RFC 768 (1980)Google Scholar
  5. 5.
    InfiniBand Trade Association.
  6. 6.
    Ajima, Y., Sumimoto, S., Shimizu, T.: Tofu: a 6D mesh/torus interconnect for exascale computers. IEEE Comput. 42(11), 30–40 (2010)Google Scholar
  7. 7.
    Ajima, Y., Inoue, T., et al.: The tofu interconnect. IEEE Micro 32(1), 21–31 (2012)CrossRefGoogle Scholar
  8. 8.
    Ajima, Y., et al.: Tofu interconnect 2: system-on-chip integration of high-performance interconnect. In: Kunkel, J.M., Ludwig, T., Meuer, H.W. (eds.) ISC 2014. LNCS, vol. 8488, pp. 498–507. Springer, Heidelberg (2014)Google Scholar
  9. 9.
    Ajima, Y., Inoue, T., et al.: The tofu interconnect 2. In: IEEE 22nd High-Performance Interconnects, pp. 57–62 (2014)Google Scholar
  10. 10.
    ARMCI – Aggregate Remote Memory Copy Interface.
  11. 11.
    GASNet Communication System.
  12. 12.
    Berkeley UPC – Unified Parallel C.
  13. 13.
  14. 14.
    UCX – Unified Communication X.
  15. 15.

Copyright information

© Springer International Publishing AG 2016

Authors and Affiliations

  • Yuichiro Ajima
    • 1
    Email author
  • Takafumi Nose
    • 1
  • Kazushige Saga
    • 1
  • Naoyuki Shida
    • 1
  • Shinji Sumimoto
    • 1
  1. 1.Fujitsu Limited/JST-CRESTKawasakiJapan

Personalised recommendations